Yan Yan

Songsong Wu, Hao Tang, Xiao-Yuan Jing et al.

In this paper, we tackle the problem of synthesizing a ground-view panorama image conditioned on a top-view aerial image, which is a challenging problem due to the large gap between the two image domains with different view-points. Instead of learning cross-view mapping in a feedforward pass, we propose a novel adversarial feedback GAN framework named PanoGAN with two key components: an adversarial feedback module and a dual branch discrimination strategy. First, the aerial image is fed into the generator to produce a target panorama image and its associated segmentation map in favor of model training with layout semantics. Second, the feature responses of the discriminator encoded by our adversarial feedback module are fed back to the generator to refine the intermediate representations, so that the generation performance is continually improved through an iterative generation process. Third, to pursue high-fidelity and semantic consistency of the generated panorama image, we propose a pixel-segmentation alignment mechanism under the dual branch discrimiantion strategy to facilitate cooperation between the generator and the discriminator. Extensive experimental results on two challenging cross-view image datasets show that PanoGAN enables high-quality panorama image generation with more convincing details than state-of-the-art approaches. The source code and trained models are available at \url{https://github.com/sswuai/PanoGAN}.

Jichao Zhang, Jingjing Chen, Hao Tang et al.

This paper proposes a gaze correction and animation method for high-resolution, unconstrained portrait images, which can be trained without the gaze angle and the head pose annotations. Common gaze-correction methods usually require annotating training data with precise gaze, and head pose information. Solving this problem using an unsupervised method remains an open problem, especially for high-resolution face images in the wild, which are not easy to annotate with gaze and head pose labels. To address this issue, we first create two new portrait datasets: CelebGaze and high-resolution CelebHQGaze. Second, we formulate the gaze correction task as an image inpainting problem, addressed using a Gaze Correction Module (GCM) and a Gaze Animation Module (GAM). Moreover, we propose an unsupervised training strategy, i.e., Synthesis-As-Training, to learn the correlation between the eye region features and the gaze angle. As a result, we can use the learned latent space for gaze animation with semantic interpolation in this space. Moreover, to alleviate both the memory and the computational costs in the training and the inference stage, we propose a Coarse-to-Fine Module (CFM) integrated with GCM and GAM. Extensive experiments validate the effectiveness of our method for both the gaze correction and the gaze animation tasks in both low and high-resolution face datasets in the wild and demonstrate the superiority of our method with respect to the state of the arts. Code is available at https://github.com/zhangqianhui/GazeAnimationV2

Xinyi Zou, Yan Yan, Jing-Hao Xue et al.

Most existing compound facial expression recognition (FER) methods rely on large-scale labeled compound expression data for training. However, collecting such data is labor-intensive and time-consuming. In this paper, we address the compound FER task in the cross-domain few-shot learning (FSL) setting, which requires only a few samples of compound expressions in the target domain. Specifically, we propose a novel cascaded decomposition network (CDNet), which cascades several learn-to-decompose modules with shared parameters based on a sequential decomposition mechanism, to obtain a transferable feature space. To alleviate the overfitting problem caused by limited base classes in our task, a partial regularization strategy is designed to effectively exploit the best of both episodic training and batch training. By training across similar tasks on multiple basic expression datasets, CDNet learns the ability of learn-to-decompose that can be easily adapted to identify unseen compound expressions. Extensive experiments on both in-the-lab and in-the-wild compound expression datasets demonstrate the superiority of our proposed CDNet against several state-of-the-art FSL methods. Code is available at: https://github.com/zouxinyi0625/CDNet.

Ye Zhu, Kyle Olszewski, Yu Wu et al.

We present Dance2Music-GAN (D2M-GAN), a novel adversarial multi-modal framework that generates complex musical samples conditioned on dance videos. Our proposed framework takes dance video frames and human body motions as input, and learns to generate music samples that plausibly accompany the corresponding input. Unlike most existing conditional music generation works that generate specific types of mono-instrumental sounds using symbolic audio representations (e.g., MIDI), and that usually rely on pre-defined musical synthesizers, in this work we generate dance music in complex styles (e.g., pop, breaking, etc.) by employing a Vector Quantized (VQ) audio representation, and leverage both its generality and high abstraction capacity of its symbolic and continuous counterparts. By performing an extensive set of experiments on multiple datasets, and following a comprehensive evaluation protocol, we assess the generative qualities of our proposal against alternatives. The attained quantitative results, which measure the music consistency, beats correspondence, and music diversity, demonstrate the effectiveness of our proposed method. Last but not least, we curate a challenging dance-music dataset of in-the-wild TikTok videos, which we use to further demonstrate the efficacy of our approach in real-world applications -- and which we hope to serve as a starting point for relevant future research.

Ye Zhu, Yu Wu, Zhiwei Deng et al.

Applying pre-trained generative denoising diffusion models (DDMs) for downstream tasks such as image semantic editing usually requires either fine-tuning DDMs or learning auxiliary editing networks in the existing literature. In this work, we present our BoundaryDiffusion method for efficient, effective and light-weight semantic control with frozen pre-trained DDMs, without learning any extra networks. As one of the first learning-free diffusion editing works, we start by seeking a comprehensive understanding of the intermediate high-dimensional latent spaces by theoretically and empirically analyzing their probabilistic and geometric behaviors in the Markov chain. We then propose to further explore the critical step for editing in the denoising trajectory that characterizes the convergence of a pre-trained DDM and introduce an automatic search method. Last but not least, in contrast to the conventional understanding that DDMs have relatively poor semantic behaviors, we prove that the critical latent space we found already exhibits semantic subspace boundaries at the generic level in unconditional DDMs, which allows us to do controllable manipulation by guiding the denoising trajectory towards the targeted boundary via a single-step operation. We conduct extensive experiments on multiple DPMs architectures (DDPM, iDDPM) and datasets (CelebA, CelebA-HQ, LSUN-church, LSUN-bedroom, AFHQ-dog) with different resolutions (64, 256), achieving superior or state-of-the-art performance in various task scenarios (image semantic editing, text-based editing, unconditional semantic control) to demonstrate the effectiveness.

Yudong Han, Liqiang Nie, Jianhua Yin et al.

Several studies have recently pointed that existing Visual Question Answering (VQA) models heavily suffer from the language prior problem, which refers to capturing superficial statistical correlations between the question type and the answer whereas ignoring the image contents. Numerous efforts have been dedicated to strengthen the image dependency by creating the delicate models or introducing the extra visual annotations. However, these methods cannot sufficiently explore how the visual cues explicitly affect the learned answer representation, which is vital for language reliance alleviation. Moreover, they generally emphasize the class-level discrimination of the learned answer representation, which overlooks the more fine-grained instance-level patterns and demands further optimization. In this paper, we propose a novel collaborative learning scheme from the viewpoint of visual perturbation calibration, which can better investigate the fine-grained visual effects and mitigate the language prior problem by learning the instance-level characteristics. Specifically, we devise a visual controller to construct two sorts of curated images with different perturbation extents, based on which the collaborative learning of intra-instance invariance and inter-instance discrimination is implemented by two well-designed discriminators. Besides, we implement the information bottleneck modulator on latent space for further bias alleviation and representation calibration. We impose our visual perturbation-aware framework to three orthodox baselines and the experimental results on two diagnostic VQA-CP benchmark datasets evidently demonstrate its effectiveness. In addition, we also justify its robustness on the balanced VQA benchmark.

Bin Xie, Hao Tang, Bin Duan et al.

Brain vessel image segmentation can be used as a promising biomarker for better prevention and treatment of different diseases. One successful approach is to consider the segmentation as an image-to-image translation task and perform a conditional Generative Adversarial Network (cGAN) to learn a transformation between two distributions. In this paper, we present a novel multi-view approach, MLP-GAN, which splits a 3D volumetric brain vessel image into three different dimensional 2D images (i.e., sagittal, coronal, axial) and then feed them into three different 2D cGANs. The proposed MLP-GAN not only alleviates the memory issue which exists in the original 3D neural networks but also retains 3D spatial information. Specifically, we utilize U-Net as the backbone for our generator and redesign the pattern of skip connection integrated with the MLP-Mixer which has attracted lots of attention recently. Our model obtains the ability to capture cross-patch information to learn global information with the MLP-Mixer. Extensive experiments are performed on the public brain vessel dataset that show our MLP-GAN outperforms other state-of-the-art methods. We release our code at https://github.com/bxie9/MLP-GAN

Xi Weng, Yan Yan, Si Chen et al.

Over the past few years, deep convolutional neural network-based methods have made great progress in semantic segmentation of street scenes. Some recent methods align feature maps to alleviate the semantic gap between them and achieve high segmentation accuracy. However, they usually adopt the feature alignment modules with the same network configuration in the decoder and thus ignore the different roles of stages of the decoder during feature aggregation, leading to a complex decoder structure. Such a manner greatly affects the inference speed. In this paper, we present a novel Stage-aware Feature Alignment Network (SFANet) based on the encoder-decoder structure for real-time semantic segmentation of street scenes. Specifically, a Stage-aware Feature Alignment module (SFA) is proposed to align and aggregate two adjacent levels of feature maps effectively. In the SFA, by taking into account the unique role of each stage in the decoder, a novel stage-aware Feature Enhancement Block (FEB) is designed to enhance spatial details and contextual information of feature maps from the encoder. In this way, we are able to address the misalignment problem with a very simple and efficient multi-branch decoder structure. Moreover, an auxiliary training strategy is developed to explicitly alleviate the multi-scale object problem without bringing additional computational costs during the inference phase. Experimental results show that the proposed SFANet exhibits a good balance between accuracy and speed for real-time semantic segmentation of street scenes. In particular, based on ResNet-18, SFANet respectively obtains 78.1% and 74.7% mean of class-wise Intersection-over-Union (mIoU) at inference speeds of 37 FPS and 96 FPS on the challenging Cityscapes and CamVid test datasets by using only a single GTX 1080Ti GPU.

Yukang Zhang, Yan Yan, Jie Li et al.

Visible-infrared person re-identification (VI-ReID), which aims to search identities across different spectra, is a challenging task due to large cross-modality discrepancy between visible and infrared images. The key to reduce the discrepancy is to filter out identity-irrelevant interference and effectively learn modality-invariant person representations. In this paper, we propose a novel Modality Restitution and Compensation Network (MRCN) to narrow the gap between the two modalities. Specifically, we first reduce the modality discrepancy by using two Instance Normalization (IN) layers. Next, to reduce the influence of IN layers on removing discriminative information and to reduce modality differences, we propose a Modality Restitution Module (MRM) and a Modality Compensation Module (MCM) to respectively distill modality-irrelevant and modality-relevant features from the removed information. Then, the modality-irrelevant features are used to restitute to the normalized visible and infrared features, while the modality-relevant features are used to compensate for the features of the other modality. Furthermore, to better disentangle the modality-relevant features and the modality-irrelevant features, we propose a novel Center-Quadruplet Causal (CQC) loss to encourage the network to effectively learn the modality-relevant features and the modality-irrelevant features. Extensive experiments are conducted to validate the superiority of our method on the challenging SYSU-MM01 and RegDB datasets. More remarkably, our method achieves 95.1% in terms of Rank-1 and 89.2% in terms of mAP on the RegDB dataset.

Xi Weng, Yan Yan, Genshun Dong et al.

Real-time semantic segmentation, which aims to achieve high segmentation accuracy at real-time inference speed, has received substantial attention over the past few years. However, many state-of-the-art real-time semantic segmentation methods tend to sacrifice some spatial details or contextual information for fast inference, thus leading to degradation in segmentation quality. In this paper, we propose a novel Deep Multi-branch Aggregation Network (called DMA-Net) based on the encoder-decoder structure to perform real-time semantic segmentation in street scenes. Specifically, we first adopt ResNet-18 as the encoder to efficiently generate various levels of feature maps from different stages of convolutions. Then, we develop a Multi-branch Aggregation Network (MAN) as the decoder to effectively aggregate different levels of feature maps and capture the multi-scale information. In MAN, a lattice enhanced residual block is designed to enhance feature representations of the network by taking advantage of the lattice structure. Meanwhile, a feature transformation block is introduced to explicitly transform the feature map from the neighboring branch before feature aggregation. Moreover, a global context block is used to exploit the global contextual information. These key components are tightly combined and jointly optimized in a unified network. Extensive experimental results on the challenging Cityscapes and CamVid datasets demonstrate that our proposed DMA-Net respectively obtains 77.0% and 73.6% mean Intersection over Union (mIoU) at the inference speed of 46.7 FPS and 119.8 FPS by only using a single NVIDIA GTX 1080Ti GPU. This shows that DMA-Net provides a good tradeoff between segmentation quality and speed for semantic segmentation in street scenes.

Jianyuan Ni, Anne H. H. Ngu, Yan Yan

Wearable sensor-based Human Action Recognition (HAR) has achieved remarkable success recently. However, the accuracy performance of wearable sensor-based HAR is still far behind the ones from the visual modalities-based system (i.e., RGB video, skeleton, and depth). Diverse input modalities can provide complementary cues and thus improve the accuracy performance of HAR, but how to take advantage of multi-modal data on wearable sensor-based HAR has rarely been explored. Currently, wearable devices, i.e., smartwatches, can only capture limited kinds of non-visual modality data. This hinders the multi-modal HAR association as it is unable to simultaneously use both visual and non-visual modality data. Another major challenge lies in how to efficiently utilize multimodal data on wearable devices with their limited computation resources. In this work, we propose a novel Progressive Skeleton-to-sensor Knowledge Distillation (PSKD) model which utilizes only time-series data, i.e., accelerometer data, from a smartwatch for solving the wearable sensor-based HAR problem. Specifically, we construct multiple teacher models using data from both teacher (human skeleton sequence) and student (time-series accelerometer data) modalities. In addition, we propose an effective progressive learning scheme to eliminate the performance gap between teacher and student models. We also designed a novel loss function called Adaptive-Confidence Semantic (ACS), to allow the student model to adaptively select either one of the teacher models or the ground-truth label it needs to mimic. To demonstrate the effectiveness of our proposed PSKD method, we conduct extensive experiments on Berkeley-MHAD, UTD-MHAD, and MMAct datasets. The results confirm that the proposed PSKD method has competitive performance compared to the previous mono sensor-based HAR methods.

Weitai Kang, Gaowen Liu, Mubarak Shah et al.

Different from Object Detection, Visual Grounding deals with detecting a bounding box for each text-image pair. This one box for each text-image data provides sparse supervision signals. Although previous works achieve impressive results, their passive utilization of annotation, i.e. the sole use of the box annotation as regression ground truth, results in a suboptimal performance. In this paper, we present SegVG, a novel method transfers the box-level annotation as Segmentation signals to provide an additional pixel-level supervision for Visual Grounding. Specifically, we propose the Multi-layer Multi-task Encoder-Decoder as the target grounding stage, where we learn a regression query and multiple segmentation queries to ground the target by regression and segmentation of the box in each decoding layer, respectively. This approach allows us to iteratively exploit the annotation as signals for both box-level regression and pixel-level segmentation. Moreover, as the backbones are typically initialized by pretrained parameters learned from unimodal tasks and the queries for both regression and segmentation are static learnable embeddings, a domain discrepancy remains among these three types of features, which impairs subsequent target grounding. To mitigate this discrepancy, we introduce the Triple Alignment module, where the query, text, and vision tokens are triangularly updated to share the same space by triple attention mechanism. Extensive experiments on five widely used datasets validate our state-of-the-art (SOTA) performance.

Qianlong Xiang, Miao Zhang, Yuzhang Shang et al.

Diffusion models (DMs) have demonstrated exceptional generative capabilities across various domains, including image, video, and so on. A key factor contributing to their effectiveness is the high quantity and quality of data used during training. However, mainstream DMs now consume increasingly large amounts of data. For example, training a Stable Diffusion model requires billions of image-text pairs. This enormous data requirement poses significant challenges for training large DMs due to high data acquisition costs and storage expenses. To alleviate this data burden, we propose a novel scenario: using existing DMs as data sources to train new DMs with any architecture. We refer to this scenario as Data-Free Knowledge Distillation for Diffusion Models (DKDM), where the generative ability of DMs is transferred to new ones in a data-free manner. To tackle this challenge, we make two main contributions. First, we introduce a DKDM objective that enables the training of new DMs via distillation, without requiring access to the data. Second, we develop a dynamic iterative distillation method that efficiently extracts time-domain knowledge from existing DMs, enabling direct retrieval of training data without the need for a prolonged generative process. To the best of our knowledge, we are the first to explore this scenario. Experimental results demonstrate that our data-free approach not only achieves competitive generative performance but also, in some instances, outperforms models trained with the entire dataset.

Zhenghao Zhao, Yuzhang Shang, Junyi Wu et al.

Deep learning has made remarkable progress recently, largely due to the availability of large, well-labeled datasets. However, the training on such datasets elevates costs and computational demands. To address this, various techniques like coreset selection, dataset distillation, and dataset quantization have been explored in the literature. Unlike traditional techniques that depend on uniform sample distributions across different classes, our research demonstrates that maintaining performance is feasible even with uneven distributions. We find that for certain classes, the variation in sample quantity has a minimal impact on performance. Inspired by this observation, an intuitive idea is to reduce the number of samples for stable classes and increase the number of samples for sensitive classes to achieve a better performance with the same sampling ratio. Then the question arises: how can we adaptively select samples from a dataset to achieve optimal performance? In this paper, we propose a novel active learning based adaptive sampling strategy, Dataset Quantization with Active Learning based Adaptive Sampling (DQAS), to optimize the sample selection. In addition, we introduce a novel pipeline for dataset quantization, utilizing feature space from the final stage of dataset quantization to generate more precise dataset bins. Our comprehensive evaluations on the multiple datasets show that our approach outperforms the state-of-the-art dataset compression methods.

Jingyu Lin, Jie Jiang, Yan Yan et al.

The prosperity of deep learning contributes to the rapid progress in scene text detection. Among all the methods with convolutional networks, segmentation-based ones have drawn extensive attention due to their superiority in detecting text instances of arbitrary shapes and extreme aspect ratios. However, the bottom-up methods are limited to the performance of their segmentation models. In this paper, we propose DPTNet (Dual-Path Transformer Network), a simple yet effective architecture to model the global and local information for the scene text detection task. We further propose a parallel design that integrates the convolutional network with a powerful self-attention mechanism to provide complementary clues between the attention path and convolutional path. Moreover, a bi-directional interaction module across the two paths is developed to provide complementary clues in the channel and spatial dimensions. We also upgrade the concentration operation by adding an extra multi-head attention layer to it. Our DPTNet achieves state-of-the-art results on the MSRA-TD500 dataset, and provides competitive results on other standard benchmarks in terms of both detection accuracy and speed.

Zhixing Hou, Yan Yan, Chengzhong Xu et al.

Place recognition or loop closure detection is one of the core components in a full SLAM system. In this paper, aiming at strengthening the relevancy of local neighboring points and the contextual dependency among global points simultaneously, we investigate the exploitation of transformer-based network for feature extraction, and propose a Hierarchical Transformer for Place Recognition (HiTPR). The HiTPR consists of four major parts: point cell generation, short-range transformer (SRT), long-range transformer (LRT) and global descriptor aggregation. Specifically, the point cloud is initially divided into a sequence of small cells by downsampling and nearest neighbors searching. In the SRT, we extract the local feature for each point cell. While in the LRT, we build the global dependency among all of the point cells in the whole point cloud. Experiments on several standard benchmarks demonstrate the superiority of the HiTPR in terms of average recall rate, achieving 93.71% at top 1% and 86.63% at top 1 on the Oxford RobotCar dataset for example.

Bin Duan, Keshav Bhandari, Gaowen Liu et al.

Optical flow estimation has been a long-lasting and fundamental problem in the computer vision community. However, despite the advances of optical flow estimation in perspective videos, the 360$^\circ$ videos counterpart remains in its infancy, primarily due to the shortage of benchmark datasets and the failure to accommodate the omnidirectional nature of 360$^\circ$ videos. We propose the first perceptually realistic 360$^\circ$ filed-of-view video benchmark dataset, namely FLOW360, with 40 different videos and 4,000 video frames. We then conduct comprehensive characteristic analysis and extensive comparisons with existing datasets, manifesting FLOW360's perceptual realism, uniqueness, and diversity. Moreover, we present a novel Siamese representation Learning framework for Omnidirectional Flow (SLOF) estimation, which is trained in a contrastive manner via a hybrid loss that combines siamese contrastive and optical flow losses. By training the model on random rotations of the input omnidirectional frames, our proposed contrastive scheme accommodates the omnidirectional nature of optical flow estimation in 360$^\circ$ videos, resulting in significantly reduced prediction errors. The learning scheme is further proven to be efficient by expanding our siamese learning scheme and omnidirectional optical flow estimation to the egocentric activity recognition task, where the classification accuracy is boosted up to $\sim$26%. To summarize, we study the optical flow estimation in 360$^\circ$ videos problem from perspectives of the benchmark dataset, learning model, and also practical application. The FLOW360 dataset and code are available at https://siamlof.github.io.

Zhiliang Wu, Hanyu Xuan, Changchang Sun et al.

Deep learning-based video inpainting has yielded promising results and gained increasing attention from researchers. Generally, these methods usually assume that the corrupted region masks of each frame are known and easily obtained. However, the annotation of these masks are labor-intensive and expensive, which limits the practical application of current methods. Therefore, we expect to relax this assumption by defining a new semi-supervised inpainting setting, making the networks have the ability of completing the corrupted regions of the whole video using the annotated mask of only one frame. Specifically, in this work, we propose an end-to-end trainable framework consisting of completion network and mask prediction network, which are designed to generate corrupted contents of the current frame using the known mask and decide the regions to be filled of the next frame, respectively. Besides, we introduce a cycle consistency loss to regularize the training parameters of these two networks. In this way, the completion network and the mask prediction network can constrain each other, and hence the overall performance of the trained model can be maximized. Furthermore, due to the natural existence of prior knowledge (e.g., corrupted contents and clear borders), current video inpainting datasets are not suitable in the context of semi-supervised video inpainting. Thus, we create a new dataset by simulating the corrupted video of real-world scenarios. Extensive experimental results are reported to demonstrate the superiority of our model in the video inpainting task. Remarkably, although our model is trained in a semi-supervised manner, it can achieve comparable performance as fully-supervised methods.

Yongzhi Huang, Hanwen Zhang, Yan Yan et al.

Large curated datasets are necessary, but annotating medical images is a time-consuming, laborious, and expensive process. Therefore, recent supervised methods are focusing on utilizing a large amount of unlabeled data. However, to do so, is a challenging task. To address this problem, we propose a new 3D Cross-Pseudo Supervision (3D-CPS) method, a semi-supervised network architecture based on nnU-Net with the Cross-Pseudo Supervision method. We design a new nnU-Net based preprocessing. In addition, we set the semi-supervised loss weights to expand linearity with each epoch to prevent the model from low-quality pseudo-labels in the early training process. Our proposed method achieves an average dice similarity coefficient (DSC) of 0.881 and an average normalized surface distance (NSD) of 0.913 on the MICCAI FLARE2022 validation set (20 cases).

Keshav Bhandari, Bin Duan, Gaowen Liu et al.

Optical flow estimation in omnidirectional videos faces two significant issues: the lack of benchmark datasets and the challenge of adapting perspective video-based methods to accommodate the omnidirectional nature. This paper proposes the first perceptually natural-synthetic omnidirectional benchmark dataset with a 360-degree field of view, FLOW360, with 40 different videos and 4,000 video frames. We conduct comprehensive characteristic analysis and comparisons between our dataset and existing optical flow datasets, which manifest perceptual realism, uniqueness, and diversity. To accommodate the omnidirectional nature, we present a novel Siamese representation Learning framework for Omnidirectional Flow (SLOF). We train our network in a contrastive manner with a hybrid loss function that combines contrastive loss and optical flow loss. Extensive experiments verify the proposed framework's effectiveness and show up to 40% performance improvement over the state-of-the-art approaches. Our FLOW360 dataset and code are available at https://siamlof.github.io/.

Zhixing Hou, Yuzhang Shang, Tian Gao et al.

Place recognition, an algorithm to recognize the re-visited places, plays the role of back-end optimization trigger in a full SLAM system. Many works equipped with deep learning tools, such as MLP, CNN, and transformer, have achieved great improvements in this research field. Point cloud transformer is one of the excellent frameworks for place recognition applied in robotics, but with large memory consumption and expensive computation, it is adverse to widely deploy the various point cloud transformer networks in mobile or embedded devices. To solve this issue, we propose a binary point cloud transformer for place recognition. As a result, a 32-bit full-precision model can be reduced to a 1-bit model with less memory occupation and faster binarized bitwise operations. To our best knowledge, this is the first binary point cloud transformer that can be deployed on mobile devices for online applications such as place recognition. Experiments on several standard benchmarks demonstrate that the proposed method can get comparable results with the corresponding full-precision transformer model and even outperform some full-precision deep learning methods. For example, the proposed method achieves 93.28% at the top @1% and 85.74% at the top @1% on the Oxford RobotCar dataset in terms of the metric of the average recall rate. Meanwhile, the size and floating point operations of the model with the same transformer structure reduce 56.1% and 34.1% respectively from original precision to binary precision.

Zhenghao Zhao, Ye Zhu, Xiaoguang Zhu et al.

Most current AI systems rely on the premise that the input visual data are sufficient to achieve competitive performance in various computer vision tasks. However, the classic task setup rarely considers the challenging, yet common practical situations where the complete visual data may be inaccessible due to various reasons (e.g., restricted view range and occlusions). To this end, we investigate a computer vision task setting with incomplete visual input data. Specifically, we exploit the Scene Graph Generation (SGG) task with various levels of visual data missingness as input. While insufficient visual input intuitively leads to performance drop, we propose to supplement the missing visions via the natural language dialog interactions to better accomplish the task objective. We design a model-agnostic Supplementary Interactive Dialog (SI-Dial) framework that can be jointly learned with most existing models, endowing the current AI systems with the ability of question-answer interactions in natural language. We demonstrate the feasibility of such a task setting with missing visual input and the effectiveness of our proposed dialog module as the supplementary information source through extensive experiments and analysis, by achieving promising performance improvement over multiple baselines.

Ruikang Chen, Yan Yan, Jing-Hao Xue et al.

Automatic X-ray prohibited item detection is vital for public safety. Existing deep learning-based methods all assume that the annotations of training X-ray images are correct. However, obtaining correct annotations is extremely hard if not impossible for large-scale X-ray images, where item overlapping is ubiquitous.As a result, X-ray images are easily contaminated with noisy annotations, leading to performance deterioration of existing methods.In this paper, we address the challenging problem of training a robust prohibited item detector under noisy annotations (including both category noise and bounding box noise) from a novel perspective of data augmentation, and propose an effective label-aware mixed patch paste augmentation method (Mix-Paste). Specifically, for each item patch, we mix several item patches with the same category label from different images and replace the original patch in the image with the mixed patch. In this way, the probability of containing the correct prohibited item within the generated image is increased. Meanwhile, the mixing process mimics item overlapping, enabling the model to learn the characteristics of X-ray images. Moreover, we design an item-based large-loss suppression (LLS) strategy to suppress the large losses corresponding to potentially positive predictions of additional items due to the mixing operation. We show the superiority of our method on X-ray datasets under noisy annotations. In addition, we evaluate our method on the noisy MS-COCO dataset to showcase its generalization ability. These results clearly indicate the great potential of data augmentation to handle noise annotations. The source code is released at https://github.com/wscds/Mix-Paste.

Yuzhang Shang, Dan Xu, Gaowen Liu et al.

Multi-task learning for dense prediction has emerged as a pivotal area in computer vision, enabling simultaneous processing of diverse yet interrelated pixel-wise prediction tasks. However, the substantial computational demands of state-of-the-art (SoTA) models often limit their widespread deployment. This paper addresses this challenge by introducing network binarization to compress resource-intensive multi-task dense predictors. Specifically, our goal is to significantly accelerate multi-task dense prediction models via Binary Neural Networks (BNNs) while maintaining and even improving model performance at the same time. To reach this goal, we propose a Binary Multi-task Dense Predictor, Bi-MTDP, and several variants of Bi-MTDP, in which a multi-task dense predictor is constructed via specified binarized modules. Our systematical analysis of this predictor reveals that performance drop from binarization is primarily caused by severe information degradation. To address this issue, we introduce a deep information bottleneck layer that enforces representations for downstream tasks satisfying Gaussian distribution in forward propagation. Moreover, we introduce a knowledge distillation mechanism to correct the direction of information flow in backward propagation. Intriguingly, one variant of Bi-MTDP outperforms full-precision (FP) multi-task dense prediction SoTAs, ARTC (CNN-based) and InvPT (ViT-Based). This result indicates that Bi-MTDP is not merely a naive trade-off between performance and efficiency, but is rather a benefit of the redundant information flow thanks to the multi-task architecture. Code is available at https://github.com/42Shawn/BiMTDP.

Hu Ding, Yan Yan, Yang Lu et al.

Most facial expression recognition (FER) models are trained on large-scale expression data with centralized learning. Unfortunately, collecting a large amount of centralized expression data is difficult in practice due to privacy concerns of facial images. In this paper, we investigate FER under the framework of personalized federated learning, which is a valuable and practical decentralized setting for real-world applications. To this end, we develop a novel uncertainty-Aware label refineMent on hYpergraphs (AMY) method. For local training, each local model consists of a backbone, an uncertainty estimation (UE) block, and an expression classification (EC) block. In the UE block, we leverage a hypergraph to model complex high-order relationships between expression samples and incorporate these relationships into uncertainty features. A personalized uncertainty estimator is then introduced to estimate reliable uncertainty weights of samples in the local client. In the EC block, we perform label propagation on the hypergraph, obtaining high-quality refined labels for retraining an expression classifier. Based on the above, we effectively alleviate heterogeneous sample uncertainty across clients and learn a robust personalized FER model in each client. Experimental results on two challenging real-world facial expression databases show that our proposed method consistently outperforms several state-of-the-art methods. This indicates the superiority of hypergraph modeling for uncertainty estimation and label refinement on the personalized federated FER task. The source code will be released at https://github.com/mobei1006/AMY.

Zhuoning Yuan, Yan Yan, Milan Sonka et al.

Deep AUC Maximization (DAM) is a new paradigm for learning a deep neural network by maximizing the AUC score of the model on a dataset. Most previous works of AUC maximization focus on the perspective of optimization by designing efficient stochastic algorithms, and studies on generalization performance of large-scale DAM on difficult tasks are missing. In this work, we aim to make DAM more practical for interesting real-world applications (e.g., medical image classification). First, we propose a new margin-based min-max surrogate loss function for the AUC score (named as AUC min-max-margin loss or simply AUC margin loss for short). It is more robust than the commonly used AUC square loss, while enjoying the same advantage in terms of large-scale stochastic optimization. Second, we conduct extensive empirical studies of our DAM method on four difficult medical image classification tasks, namely (i) classification of chest x-ray images for identifying many threatening diseases, (ii) classification of images of skin lesions for identifying melanoma, (iii) classification of mammogram for breast cancer screening, and (iv) classification of microscopic images for identifying tumor tissue. Our studies demonstrate that the proposed DAM method improves the performance of optimizing cross-entropy loss by a large margin, and also achieves better performance than optimizing the existing AUC square loss on these medical image classification tasks. Specifically, our DAM method has achieved the 1st place on Stanford CheXpert competition on Aug. 31, 2020. To the best of our knowledge, this is the first work that makes DAM succeed on large-scale medical image datasets. We also conduct extensive ablation studies to demonstrate the advantages of the new AUC margin loss over the AUC square loss on benchmark datasets. The proposed method is implemented in our open-sourced library LibAUC (www.libauc.org).

Jichao Zhang, Jingjing Chen, Hao Tang et al.

In this paper we address the problem of unsupervised gaze correction in the wild, presenting a solution that works without the need for precise annotations of the gaze angle and the head pose. We have created a new dataset called CelebAGaze, which consists of two domains X, Y, where the eyes are either staring at the camera or somewhere else. Our method consists of three novel modules: the Gaze Correction module (GCM), the Gaze Animation module (GAM), and the Pretrained Autoencoder module (PAM). Specifically, GCM and GAM separately train a dual in-painting network using data from the domain $X$ for gaze correction and data from the domain $Y$ for gaze animation. Additionally, a Synthesis-As-Training method is proposed when training GAM to encourage the features encoded from the eye region to be correlated with the angle information, resulting in a gaze animation which can be achieved by interpolation in the latent space. To further preserve the identity information~(e.g., eye shape, iris color), we propose the PAM with an Autoencoder, which is based on Self-Supervised mirror learning where the bottleneck features are angle-invariant and which works as an extra input to the dual in-painting models. Extensive experiments validate the effectiveness of the proposed method for gaze correction and gaze animation in the wild and demonstrate the superiority of our approach in producing more compelling results than state-of-the-art baselines. Our code, the pretrained models and the supplementary material are available at: https://github.com/zhangqianhui/GazeAnimation.

Hao Tang, Dan Xu, Yan Yan et al.

In this paper, we address the task of semantic-guided scene generation. One open challenge in scene generation is the difficulty of the generation of small objects and detailed local texture, which has been widely observed in global image-level generation methods. To tackle this issue, in this work we consider learning the scene generation in a local context, and correspondingly design a local class-specific generative network with semantic maps as a guidance, which separately constructs and learns sub-generators concentrating on the generation of different classes, and is able to provide more scene details. To learn more discriminative class-specific feature representations for the local generation, a novel classification module is also proposed. To combine the advantage of both the global image-level and the local class-specific generation, a joint generation network is designed with an attention fusion module and a dual-discriminator structure embedded. Extensive experiments on two scene image generation tasks show superior generation performance of the proposed model. The state-of-the-art results are established by large margins on both tasks and on challenging public benchmarks. The source code and trained models are available at https://github.com/Ha0Tang/LGGAN.

Qi Qi, Yan Yan, Xiaoyu Wang et al.

Deep metric learning (DML) has received much attention in deep learning due to its wide applications in computer vision. Previous studies have focused on designing complicated losses and hard example mining methods, which are mostly heuristic and lack of theoretical understanding. In this paper, we cast DML as a simple pairwise binary classification problem that classifies a pair of examples as similar or dissimilar. It identifies the most critical issue in this problem--imbalanced data pairs. To tackle this issue, we propose a simple and effective framework to sample pairs in a batch of data for updating the model. The key to this framework is to define a robust loss for all pairs over a mini-batch of data, which is formulated by distributionally robust optimization. The flexibility in constructing the uncertainty decision set of the dual variable allows us to recover state-of-the-art complicated losses and also to induce novel variants. Empirical studies on several benchmark data sets demonstrate that our simple and effective method outperforms the state-of-the-art results. Codes are available at: https://github.com/qiqi-helloworld/A-Simple-and-Effective-Framework-for-Pairewise-Distance-Metric-Learning

Bin Duan, Wei Wang, Hao Tang et al.

Since we were babies, we intuitively develop the ability to correlate the input from different cognitive sensors such as vision, audio, and text. However, in machine learning, this cross-modal learning is a nontrivial task because different modalities have no homogeneous properties. Previous works discover that there should be bridges among different modalities. From neurology and psychology perspective, humans have the capacity to link one modality with another one, e.g., associating a picture of a bird with the only hearing of its singing and vice versa. Is it possible for machine learning algorithms to recover the scene given the audio signal? In this paper, we propose a novel Cascade Attention-Guided Residue GAN (CAR-GAN), aiming at reconstructing the scenes given the corresponding audio signals. Particularly, we present a residue module to mitigate the gap between different modalities progressively. Moreover, a cascade attention guided network with a novel classification loss function is designed to tackle the cross-modal learning task. Our model keeps the consistency in high-level semantic label domain and is able to balance two different modalities. The experimental results demonstrate that our model achieves the state-of-the-art cross-modal audio-visual generation on the challenging Sub-URMP dataset. Code will be available at https://github.com/tuffr5/CAR-GAN.

Jichao Zhang, Meng Sun, Jingjing Chen et al.

Gaze correction aims to redirect the person's gaze into the camera by manipulating the eye region, and it can be considered as a specific image resynthesis problem. Gaze correction has a wide range of applications in real life, such as taking a picture with staring at the camera. In this paper, we propose a novel method that is based on the inpainting model to learn from the face image to fill in the missing eye regions with new contents representing corrected eye gaze. Moreover, our model does not require the training dataset labeled with the specific head pose and eye angle information, thus, the training data is easy to collect. To retain the identity information of the eye region in the original input, we propose a self-guided pretrained model to learn the angle-invariance feature. Experiments show our model achieves very compelling gaze-corrected results in the wild dataset which is collected from the website and will be introduced in details. Code is available at https://github.com/zhangqianhui/GazeCorrection.

Hao Tang, Dan Xu, Nicu Sebe et al.

Cross-view image translation is challenging because it involves images with drastically different views and severe deformation. In this paper, we propose a novel approach named Multi-Channel Attention SelectionGAN (SelectionGAN) that makes it possible to generate images of natural scenes in arbitrary viewpoints, based on an image of the scene and a novel semantic map. The proposed SelectionGAN explicitly utilizes the semantic information and consists of two stages. In the first stage, the condition image and the target semantic map are fed into a cycled semantic-guided generation network to produce initial coarse results. In the second stage, we refine the initial results by using a multi-channel attention selection mechanism. Moreover, uncertainty maps automatically learned from attentions are used to guide the pixel loss for better network optimization. Extensive experiments on Dayton, CVUSA and Ego2Top datasets show that our model is able to generate significantly better results than the state-of-the-art methods. The source code, data and trained models are available at https://github.com/Ha0Tang/SelectionGAN.

Hao Tang, Dan Xu, Nicu Sebe et al.

The state-of-the-art approaches in Generative Adversarial Networks (GANs) are able to learn a mapping function from one image domain to another with unpaired image data. However, these methods often produce artifacts and can only be able to convert low-level information, but fail to transfer high-level semantic part of images. The reason is mainly that generators do not have the ability to detect the most discriminative semantic part of images, which thus makes the generated images with low-quality. To handle the limitation, in this paper we propose a novel Attention-Guided Generative Adversarial Network (AGGAN), which can detect the most discriminative semantic object and minimize changes of unwanted part for semantic manipulation problems without using extra data and models. The attention-guided generators in AGGAN are able to produce attention masks via a built-in attention mechanism, and then fuse the input image with the attention mask to obtain a target image with high-quality. Moreover, we propose a novel attention-guided discriminator which only considers attended regions. The proposed AGGAN is trained by an end-to-end fashion with an adversarial loss, cycle-consistency loss, pixel loss and attention loss. Both qualitative and quantitative results demonstrate that our approach is effective to generate sharper and more accurate images than existing models. The code is available at https://github.com/Ha0Tang/AttentionGAN.

Hao Tang, Wei Wang, Dan Xu et al.

Hand gesture-to-gesture translation in the wild is a challenging task since hand gestures can have arbitrary poses, sizes, locations and self-occlusions. Therefore, this task requires a high-level understanding of the mapping between the input source gesture and the output target gesture. To tackle this problem, we propose a novel hand Gesture Generative Adversarial Network (GestureGAN). GestureGAN consists of a single generator $G$ and a discriminator $D$, which takes as input a conditional hand image and a target hand skeleton image. GestureGAN utilizes the hand skeleton information explicitly, and learns the gesture-to-gesture mapping through two novel losses, the color loss and the cycle-consistency loss. The proposed color loss handles the issue of "channel pollution" while back-propagating the gradients. In addition, we present the Fréchet ResNet Distance (FRD) to evaluate the quality of generated images. Extensive experiments on two widely used benchmark datasets demonstrate that the proposed GestureGAN achieves state-of-the-art performance on the unconstrained hand gesture-to-gesture translation task. Meanwhile, the generated images are in high-quality and are photo-realistic, allowing them to be used as data augmentation to improve the performance of a hand gesture classifier. Our model and code are available at https://github.com/Ha0Tang/GestureGAN.

Xuanyi Dong, Yan Yan, Wanli Ouyang et al.

Recent advances in facial landmark detection achieve success by learning discriminative features from rich deformation of face shapes and poses. Besides the variance of faces themselves, the intrinsic variance of image styles, e.g., grayscale vs. color images, light vs. dark, intense vs. dull, and so on, has constantly been overlooked. This issue becomes inevitable as increasing web images are collected from various sources for training neural networks. In this work, we propose a style-aggregated approach to deal with the large intrinsic variance of image styles for facial landmark detection. Our method transforms original face images to style-aggregated images by a generative adversarial module. The proposed scheme uses the style-aggregated image to maintain face images that are more robust to environmental changes. Then the original face images accompanying with style-aggregated ones play a duet to train a landmark detector which is complementary to each other. In this way, for each face, our method takes two images as input, i.e., one in its original style and the other in the aggregated style. In experiments, we observe that the large variance of image styles would degenerate the performance of facial landmark detectors. Moreover, we show the robustness of our method to the large variance of image styles by comparing to a variant of our approach, in which the generative adversarial module is removed, and no style-aggregated images are used. Our approach is demonstrated to perform well when compared with state-of-the-art algorithms on benchmark datasets AFLW and 300-W. Code is publicly available on GitHub: https://github.com/D-X-Y/SAN

Haoyu Sun, Yan Yan

Due to limited budgets allocated for road maintenance projects in various countries, road management departments face difficulties in making scientific maintenance decisions. This paper aims to provide road management departments with more scientific decision tools and evidence. The framework proposed in this paper mainly has the following four innovative points: 1) Predicting pavement performance deterioration levels of road sections as decision basis rather than accurately predicting specific indicator values; 2) Determining maintenance route priorities based on multiple factors; 3) Making maintenance plan decisions by establishing deep reinforcement learning models to formulate predictive strategies based on past maintenance performance evaluations, while considering both technical and management indicators; 4) Determining repair section priorities according to actual and suggested repair effects. By resolving these four issues, the framework can make intelligent decisions regarding optimal maintenance plans and sections, taking into account limited funds and historical maintenance management experiences.

Yuzhang Shang, Zhihang Yuan, Yan Yan

Dataset distillation (DD) aims to synthesize a small dataset whose test performance is comparable to a full dataset using the same model. State-of-the-art (SoTA) methods optimize synthetic datasets primarily by matching heuristic indicators extracted from two networks: one from real data and one from synthetic data (see Fig.1, Left), such as gradients and training trajectories. DD is essentially a compression problem that emphasizes maximizing the preservation of information contained in the data. We argue that well-defined metrics which measure the amount of shared information between variables in information theory are necessary for success measurement but are never considered by previous works. Thus, we introduce mutual information (MI) as the metric to quantify the shared information between the synthetic and the real datasets, and devise MIM4DD numerically maximizing the MI via a newly designed optimizable objective within a contrastive learning framework to update the synthetic dataset. Specifically, we designate the samples in different datasets that share the same labels as positive pairs and vice versa negative pairs. Then we respectively pull and push those samples in positive and negative pairs into contrastive space via minimizing NCE loss. As a result, the targeted MI can be transformed into a lower bound represented by feature maps of samples, which is numerically feasible. Experiment results show that MIM4DD can be implemented as an add-on module to existing SoTA DD methods.

Jianyuan Ni, Hao Tang, Syed Tousiful Haque et al.

The combination of increased life expectancy and falling birth rates is resulting in an aging population. Wearable Sensor-based Human Activity Recognition (WSHAR) emerges as a promising assistive technology to support the daily lives of older individuals, unlocking vast potential for human-centric applications. However, recent surveys in WSHAR have been limited, focusing either solely on deep learning approaches or on a single sensor modality. In real life, our human interact with the world in a multi-sensory way, where diverse information sources are intricately processed and interpreted to accomplish a complex and unified sensing system. To give machines similar intelligence, multimodal machine learning, which merges data from various sources, has become a popular research area with recent advancements. In this study, we present a comprehensive survey from a novel perspective on how to leverage multimodal learning to WSHAR domain for newcomers and researchers. We begin by presenting the recent sensor modalities as well as deep learning approaches in HAR. Subsequently, we explore the techniques used in present multimodal systems for WSHAR. This includes inter-multimodal systems which utilize sensor modalities from both visual and non-visual systems and intra-multimodal systems that simply take modalities from non-visual systems. After that, we focus on current multimodal learning approaches that have applied to solve some of the challenges existing in WSHAR. Specifically, we make extra efforts by connecting the existing multimodal literature from other domains, such as computer vision and natural language processing, with current WSHAR area. Finally, we identify the corresponding challenges and potential research direction in current WSHAR area for further improvement.

Zekai Li, Xinhao Zhong, Samir Khaki et al.

In recent years, dataset distillation has provided a reliable solution for data compression, where models trained on the resulting smaller synthetic datasets achieve performance comparable to those trained on the original datasets. To further improve the performance of synthetic datasets, various training pipelines and optimization objectives have been proposed, greatly advancing the field of dataset distillation. Recent decoupled dataset distillation methods introduce soft labels and stronger data augmentation during the post-evaluation phase and scale dataset distillation up to larger datasets (e.g., ImageNet-1K). However, this raises a question: Is accuracy still a reliable metric to fairly evaluate dataset distillation methods? Our empirical findings suggest that the performance improvements of these methods often stem from additional techniques rather than the inherent quality of the images themselves, with even randomly sampled images achieving superior results. Such misaligned evaluation settings severely hinder the development of DD. Therefore, we propose DD-Ranking, a unified evaluation framework, along with new general evaluation metrics to uncover the true performance improvements achieved by different methods. By refocusing on the actual information enhancement of distilled datasets, DD-Ranking provides a more comprehensive and fair evaluation standard for future research advancements.

Hooman Shahrokhi, Devjeet Raj Roy, Yan Yan et al.

We consider the problem of generating valid and small prediction sets by sampling outputs (e.g., software code and natural language text) from a black-box deep generative model for a given input (e.g., textual prompt). The validity of a prediction set is determined by a user-defined binary admissibility function depending on the target application. For example, requiring at least one program in the set to pass all test cases in code generation application. To address this problem, we develop a simple and effective conformal inference algorithm referred to as Generative Prediction Sets (GPS). Given a set of calibration examples and black-box access to a deep generative model, GPS can generate prediction sets with provable guarantees. The key insight behind GPS is to exploit the inherent structure within the distribution over the minimum number of samples needed to obtain an admissible output to develop a simple conformal regression approach over the minimum number of samples. Experiments on multiple datasets for code and math word problems using different large language models demonstrate the efficacy of GPS over state-of-the-art methods.

Bin Xie, Hao Tang, Dawen Cai et al.

Segment Anything Model (SAM) has demonstrated impressive zero-shot performance and brought a range of unexplored capabilities to natural image segmentation tasks. However, as a very important branch of image segmentation, the performance of SAM remains uncertain when applied to medical image segmentation due to the significant differences between natural images and medical images. Meanwhile, it is harsh to meet the SAM's requirements of extra prompts provided, such as points or boxes to specify medical regions. In this paper, we propose a novel self-prompt SAM adaptation framework for medical image segmentation, named Self-Prompt-SAM. We design a multi-scale prompt generator combined with the image encoder in SAM to generate auxiliary masks. Then, we use the auxiliary masks to generate bounding boxes as box prompts and use Distance Transform to select the most central points as point prompts. Meanwhile, we design a 3D depth-fused adapter (DfusedAdapter) and inject the DFusedAdapter into each transformer in the image encoder and mask decoder to enable pre-trained 2D SAM models to extract 3D information and adapt to 3D medical images. Extensive experiments demonstrate that our method achieves state-of-the-art performance and outperforms nnUNet by 2.3% on AMOS2022, 1.6% on ACDCand 0.5% on Synapse datasets.

Bin Xie, Hao Tang, Yan Yan et al.

Segment Anything Model 2 (SAM 2), a prompt-driven foundation model extending SAM to both image and video domains, has shown superior zero-shot performance compared to its predecessor. Building on SAM's success in medical image segmentation, SAM 2 presents significant potential for further advancement. However, similar to SAM, SAM 2 is limited by its output of binary masks, inability to infer semantic labels, and dependence on precise prompts for the target object area. Additionally, direct application of SAM and SAM 2 to medical image segmentation tasks yields suboptimal results. In this paper, we explore the upper performance limit of SAM 2 using custom fine-tuning adapters, achieving a Dice Similarity Coefficient (DSC) of 92.30% on the BTCV dataset, surpassing the state-of-the-art nnUNet by 12%. Following this, we address the prompt dependency by investigating various prompt generators. We introduce a UNet to autonomously generate predicted masks and bounding boxes, which serve as input to SAM 2. Subsequent dual-stage refinements by SAM 2 further enhance performance. Extensive experiments show that our method achieves state-of-the-art results on the AMOS2022 dataset, with a Dice improvement of 2.9% compared to nnUNet, and outperforms nnUNet by 6.4% on the BTCV dataset.

Haoxuan Wang, Zhenghao Zhao, Junyi Wu et al.

The recent introduction of diffusion models in dataset distillation has shown promising potential in creating compact surrogate datasets for large, high-resolution target datasets, offering improved efficiency and performance over traditional bi-level/uni-level optimization methods. However, current diffusion-based dataset distillation approaches overlook the evaluation process and exhibit two critical inconsistencies in the distillation process: (1) Objective Inconsistency, where the distillation process diverges from the evaluation objective, and (2) Condition Inconsistency, leading to mismatches between generated images and their corresponding conditions. To resolve these issues, we introduce Condition-aware Optimization with Objective-guided Sampling (CaO$_2$), a two-stage diffusion-based framework that aligns the distillation process with the evaluation objective. The first stage employs a probability-informed sample selection pipeline, while the second stage refines the corresponding latent representations to improve conditional likelihood. CaO$_2$ achieves state-of-the-art performance on ImageNet and its subsets, surpassing the best-performing baselines by an average of 2.3% accuracy.

Bin Xie, Hao Tang, Bin Duan et al.

Segment Anything Model (SAM), a prompt-driven foundation model for natural image segmentation, has demonstrated impressive zero-shot performance. However, SAM does not work when directly applied to medical image segmentation, since SAM lacks the ability to predict semantic labels, requires additional prompts, and presents suboptimal performance. Following the above issues, we propose MaskSAM, a novel mask classification prompt-free SAM adaptation framework for medical image segmentation. We design a prompt generator combined with the image encoder in SAM to generate a set of auxiliary classifier tokens, auxiliary binary masks, and auxiliary bounding boxes. Each pair of auxiliary mask and box prompts can solve the requirements of extra prompts. The semantic label prediction can be addressed by the sum of the auxiliary classifier tokens and the learnable global classifier tokens in the mask decoder of SAM. Meanwhile, we design a 3D depth-convolution adapter for image embeddings and a 3D depth-MLP adapter for prompt embeddings to efficiently fine-tune SAM. Our method achieves state-of-the-art performance on AMOS2022, 90.52% Dice, which improved by 2.7% compared to nnUNet. Our method surpasses nnUNet by 1.7% on ACDC and 1.0% on Synapse datasets.

Yukang Zhang, Yang Lu, Yan Yan et al.

The key of visible-infrared person re-identification (VIReID) lies in how to minimize the modality discrepancy between visible and infrared images. Existing methods mainly exploit the spatial information while ignoring the discriminative frequency information. To address this issue, this paper aims to reduce the modality discrepancy from the frequency domain perspective. Specifically, we propose a novel Frequency Domain Nuances Mining (FDNM) method to explore the cross-modality frequency domain information, which mainly includes an amplitude guided phase (AGP) module and an amplitude nuances mining (ANM) module. These two modules are mutually beneficial to jointly explore frequency domain visible-infrared nuances, thereby effectively reducing the modality discrepancy in the frequency domain. Besides, we propose a center-guided nuances mining loss to encourage the ANM module to preserve discriminative identity information while discovering diverse cross-modality nuances. Extensive experiments show that the proposed FDNM has significant advantages in improving the performance of VIReID. Specifically, our method outperforms the second-best method by 5.2\% in Rank-1 accuracy and 5.8\% in mAP on the SYSU-MM01 dataset under the indoor search mode, respectively. Besides, we also validate the effectiveness and generalization of our method on the challenging visible-infrared face recognition task. \textcolor{magenta}{The code will be available.}

Ziqiang Zhang, Yan Yan, Jing-Hao Xue et al.

Most existing GAN inversion methods either achieve accurate reconstruction but lack editability or offer strong editability at the cost of fidelity. Hence, how to balance the distortioneditability trade-off is a significant challenge for GAN inversion. To address this challenge, we introduce a novel spatial-contextual discrepancy information compensationbased GAN-inversion method (SDIC), which consists of a discrepancy information prediction network (DIPN) and a discrepancy information compensation network (DICN). SDIC follows a "compensate-and-edit" paradigm and successfully bridges the gap in image details between the original image and the reconstructed/edited image. On the one hand, DIPN encodes the multi-level spatial-contextual information of the original and initial reconstructed images and then predicts a spatial-contextual guided discrepancy map with two hourglass modules. In this way, a reliable discrepancy map that models the contextual relationship and captures finegrained image details is learned. On the other hand, DICN incorporates the predicted discrepancy information into both the latent code and the GAN generator with different transformations, generating high-quality reconstructed/edited images. This effectively compensates for the loss of image details during GAN inversion. Both quantitative and qualitative experiments demonstrate that our proposed method achieves the excellent distortion-editability trade-off at a fast inference speed for both image inversion and editing tasks.

Zhenghao Zhao, Haoxuan Wang, Junyi Wu et al.

Dataset distillation aims to synthesize a small dataset from a large dataset, enabling the model trained on it to perform well on the original dataset. With the blooming of large language models and multimodal large language models, the importance of multimodal datasets, particularly image-text datasets, has grown significantly. However, existing multimodal dataset distillation methods are constrained by the Matching Training Trajectories algorithm, which significantly increases the computing resource requirement, and takes days to process the distillation. In this work, we introduce EDGE, a generative distillation method for efficient multimodal dataset distillation. Specifically, we identify two key challenges of distilling multimodal datasets with generative models: 1) The lack of correlation between generated images and captions. 2) The lack of diversity among generated samples. To address the aforementioned issues, we propose a novel generative model training workflow with a bi-directional contrastive loss and a diversity loss. Furthermore, we propose a caption synthesis strategy to further improve text-to-image retrieval performance by introducing more text information. Our method is evaluated on Flickr30K, COCO, and CC3M datasets, demonstrating superior performance and efficiency compared to existing approaches. Notably, our method achieves results 18x faster than the state-of-the-art method.

Danhui Chen, Ziquan Liu, Chuxi Yang et al.

Pixel-level vision tasks, such as semantic segmentation, require extensive and high-quality annotated data, which is costly to obtain. Semi-supervised semantic segmentation (SSSS) has emerged as a solution to alleviate the labeling burden by leveraging both labeled and unlabeled data through self-training techniques. Meanwhile, the advent of foundational segmentation models pre-trained on massive data, has shown the potential to generalize across domains effectively. This work explores whether a foundational segmentation model can address label scarcity in the pixel-level vision task as an annotator for unlabeled images. Specifically, we investigate the efficacy of using SEEM, a Segment Anything Model (SAM) variant fine-tuned for textual input, to generate predictive masks for unlabeled data. To address the shortcomings of using SEEM-generated masks as supervision, we propose ConformalSAM, a novel SSSS framework which first calibrates the foundation model using the target domain's labeled data and then filters out unreliable pixel labels of unlabeled data so that only high-confidence labels are used as supervision. By leveraging conformal prediction (CP) to adapt foundation models to target data through uncertainty calibration, ConformalSAM exploits the strong capability of the foundational segmentation model reliably which benefits the early-stage learning, while a subsequent self-reliance training strategy mitigates overfitting to SEEM-generated masks in the later training stage. Our experiment demonstrates that, on three standard benchmarks of SSSS, ConformalSAM achieves superior performance compared to recent SSSS methods and helps boost the performance of those methods as a plug-in.

Yuanjie Shi, Subhankar Ghosh, Taha Belkhouja et al.

Conformal prediction (CP) is an emerging uncertainty quantification framework that allows us to construct a prediction set to cover the true label with a pre-specified marginal or conditional probability. Although the valid coverage guarantee has been extensively studied for classification problems, CP often produces large prediction sets which may not be practically useful. This issue is exacerbated for the setting of class-conditional coverage on imbalanced classification tasks with many and/or imbalanced classes. This paper proposes the Rank Calibrated Class-conditional CP (RC3P) algorithm to reduce the prediction set sizes to achieve class-conditional coverage, where the valid coverage holds for each class. In contrast to the standard class-conditional CP (CCP) method that uniformly thresholds the class-wise conformity score for each class, the augmented label rank calibration step allows RC3P to selectively iterate this class-wise thresholding subroutine only for a subset of classes whose class-wise top-k error is small. We prove that agnostic to the classifier and data distribution, RC3P achieves class-wise coverage. We also show that RC3P reduces the size of prediction sets compared to the CCP method. Comprehensive experiments on multiple real-world datasets demonstrate that RC3P achieves class-wise coverage and 26.25% reduction in prediction set sizes on average.

Yuheng Shi, Zehao Huang, Yan Yan et al.

Time-to-Contact (TTC) estimation is a critical task for assessing collision risk and is widely used in various driver assistance and autonomous driving systems. The past few decades have witnessed development of related theories and algorithms. The prevalent learning-based methods call for a large-scale TTC dataset in real-world scenarios. In this work, we present a large-scale object oriented TTC dataset in the driving scene for promoting the TTC estimation by a monocular camera. To collect valuable samples and make data with different TTC values relatively balanced, we go through thousands of hours of driving data and select over 200K sequences with a preset data distribution. To augment the quantity of small TTC cases, we also generate clips using the latest Neural rendering methods. Additionally, we provide several simple yet effective TTC estimation baselines and evaluate them extensively on the proposed dataset to demonstrate their effectiveness. The proposed dataset is publicly available at https://open-dataset.tusen.ai/TSTTC.