Yu Liu

Jihao Liu, Boxiao Liu, Hang Zhou et al.

CutMix is a popular augmentation technique commonly used for training modern convolutional and transformer vision networks. It was originally designed to encourage Convolution Neural Networks (CNNs) to focus more on an image's global context instead of local information, which greatly improves the performance of CNNs. However, we found it to have limited benefits for transformer-based architectures that naturally have a global receptive field. In this paper, we propose a novel data augmentation technique TokenMix to improve the performance of vision transformers. TokenMix mixes two images at token level via partitioning the mixing region into multiple separated parts. Besides, we show that the mixed learning target in CutMix, a linear combination of a pair of the ground truth labels, might be inaccurate and sometimes counter-intuitive. To obtain a more suitable target, we propose to assign the target score according to the content-based neural activation maps of the two images from a pre-trained teacher model, which does not need to have high performance. With plenty of experiments on various vision transformer architectures, we show that our proposed TokenMix helps vision transformers focus on the foreground area to infer the classes and enhances their robustness to occlusion, with consistent performance gains. Notably, we improve DeiT-T/S/B with +1% ImageNet top-1 accuracy. Besides, TokenMix enjoys longer training, which achieves 81.2% top-1 accuracy on ImageNet with DeiT-S trained for 400 epochs. Code is available at https://github.com/Sense-X/TokenMix.

Dongzhi Jiang, Renrui Zhang, Ziyu Guo et al. · pku

The advent of Large Language Models (LLMs) has paved the way for AI search engines, e.g., SearchGPT, showcasing a new paradigm in human-internet interaction. However, most current AI search engines are limited to text-only settings, neglecting the multimodal user queries and the text-image interleaved nature of website information. Recently, Large Multimodal Models (LMMs) have made impressive strides. Yet, whether they can function as AI search engines remains under-explored, leaving the potential of LMMs in multimodal search an open question. To this end, we first design a delicate pipeline, MMSearch-Engine, to empower any LMMs with multimodal search capabilities. On top of this, we introduce MMSearch, a comprehensive evaluation benchmark to assess the multimodal search performance of LMMs. The curated dataset contains 300 manually collected instances spanning 14 subfields, which involves no overlap with the current LMMs' training data, ensuring the correct answer can only be obtained within searching. By using MMSearch-Engine, the LMMs are evaluated by performing three individual tasks (requery, rerank, and summarization), and one challenging end-to-end task with a complete searching process. We conduct extensive experiments on closed-source and open-source LMMs. Among all tested models, GPT-4o with MMSearch-Engine achieves the best results, which surpasses the commercial product, Perplexity Pro, in the end-to-end task, demonstrating the effectiveness of our proposed pipeline. We further present error analysis to unveil current LMMs still struggle to fully grasp the multimodal search tasks, and conduct ablation study to indicate the potential of scaling test-time computation for AI search engine. We hope MMSearch may provide unique insights to guide the future development of multimodal AI search engine. Project Page: https://mmsearch.github.io

Zhuofan Zong, Guanglu Song, Yu Liu

In this paper, we provide the observation that too few queries assigned as positive samples in DETR with one-to-one set matching leads to sparse supervision on the encoder's output which considerably hurt the discriminative feature learning of the encoder and vice visa for attention learning in the decoder. To alleviate this, we present a novel collaborative hybrid assignments training scheme, namely $\mathcal{C}$o-DETR, to learn more efficient and effective DETR-based detectors from versatile label assignment manners. This new training scheme can easily enhance the encoder's learning ability in end-to-end detectors by training the multiple parallel auxiliary heads supervised by one-to-many label assignments such as ATSS and Faster RCNN. In addition, we conduct extra customized positive queries by extracting the positive coordinates from these auxiliary heads to improve the training efficiency of positive samples in the decoder. In inference, these auxiliary heads are discarded and thus our method introduces no additional parameters and computational cost to the original detector while requiring no hand-crafted non-maximum suppression (NMS). We conduct extensive experiments to evaluate the effectiveness of the proposed approach on DETR variants, including DAB-DETR, Deformable-DETR, and DINO-Deformable-DETR. The state-of-the-art DINO-Deformable-DETR with Swin-L can be improved from 58.5% to 59.5% AP on COCO val. Surprisingly, incorporated with ViT-L backbone, we achieve 66.0% AP on COCO test-dev and 67.9% AP on LVIS val, outperforming previous methods by clear margins with much fewer model sizes. Codes are available at \url{https://github.com/Sense-X/Co-DETR}.

Linjiang Huang, Kaixin Lu, Guanglu Song et al.

In this paper, we present a novel training scheme, namely Teach-DETR, to learn better DETR-based detectors from versatile teacher detectors. We show that the predicted boxes from teacher detectors are effective medium to transfer knowledge of teacher detectors, which could be either RCNN-based or DETR-based detectors, to train a more accurate and robust DETR model. This new training scheme can easily incorporate the predicted boxes from multiple teacher detectors, each of which provides parallel supervisions to the student DETR. Our strategy introduces no additional parameters and adds negligible computational cost to the original detector during training. During inference, Teach-DETR brings zero additional overhead and maintains the merit of requiring no non-maximum suppression. Extensive experiments show that our method leads to consistent improvement for various DETR-based detectors. Specifically, we improve the state-of-the-art detector DINO with Swin-Large backbone, 4 scales of feature maps and 36-epoch training schedule, from 57.8% to 58.9% in terms of mean average precision on MSCOCO 2017 validation set. Code will be available at https://github.com/LeonHLJ/Teach-DETR.

Lianghua Huang, Di Chen, Yu Liu et al.

Recent large-scale generative models learned on big data are capable of synthesizing incredible images yet suffer from limited controllability. This work offers a new generation paradigm that allows flexible control of the output image, such as spatial layout and palette, while maintaining the synthesis quality and model creativity. With compositionality as the core idea, we first decompose an image into representative factors, and then train a diffusion model with all these factors as the conditions to recompose the input. At the inference stage, the rich intermediate representations work as composable elements, leading to a huge design space (i.e., exponentially proportional to the number of decomposed factors) for customizable content creation. It is noteworthy that our approach, which we call Composer, supports various levels of conditions, such as text description as the global information, depth map and sketch as the local guidance, color histogram for low-level details, etc. Besides improving controllability, we confirm that Composer serves as a general framework and facilitates a wide range of classical generative tasks without retraining. Code and models will be made available.

Zeyue Xue, Jianming Liang, Guanglu Song et al.

Training a large-scale deep neural network in a large-scale dataset is challenging and time-consuming. The recent breakthrough of large-batch optimization is a promising way to tackle this challenge. However, although the current advanced algorithms such as LARS and LAMB succeed in classification models, the complicated pipelines of dense visual predictions such as object detection and segmentation still suffer from the heavy performance drop in the large-batch training regime. To address this challenge, we propose a simple yet effective algorithm, named Adaptive Gradient Variance Modulator (AGVM), which can train dense visual predictors with very large batch size, enabling several benefits more appealing than prior arts. Firstly, AGVM can align the gradient variances between different modules in the dense visual predictors, such as backbone, feature pyramid network (FPN), detection, and segmentation heads. We show that training with a large batch size can fail with the gradient variances misaligned among them, which is a phenomenon primarily overlooked in previous work. Secondly, AGVM is a plug-and-play module that generalizes well to many different architectures (e.g., CNNs and Transformers) and different tasks (e.g., object detection, instance segmentation, semantic segmentation, and panoptic segmentation). It is also compatible with different optimizers (e.g., SGD and AdamW). Thirdly, a theoretical analysis of AGVM is provided. Extensive experiments on the COCO and ADE20K datasets demonstrate the superiority of AGVM. For example, it can train Faster R-CNN+ResNet50 in 4 minutes without losing performance. AGVM enables training an object detector with one billion parameters in just 3.5 hours, reducing the training time by 20.9x, whilst achieving 62.2 mAP on COCO. The deliverables are released at https://github.com/Sense-X/AGVM.

Jianming Liang, Guanglu Song, Biao Leng et al.

We present a conceptually simple, flexible, and universal visual perception head for variant visual tasks, e.g., classification, object detection, instance segmentation and pose estimation, and different frameworks, such as one-stage or two-stage pipelines. Our approach effectively identifies an object in an image while simultaneously generating a high-quality bounding box or contour-based segmentation mask or set of keypoints. The method, called UniHead, views different visual perception tasks as the dispersible points learning via the transformer encoder architecture. Given a fixed spatial coordinate, UniHead adaptively scatters it to different spatial points and reasons about their relations by transformer encoder. It directly outputs the final set of predictions in the form of multiple points, allowing us to perform different visual tasks in different frameworks with the same head design. We show extensive evaluations on ImageNet classification and all three tracks of the COCO suite of challenges, including object detection, instance segmentation and pose estimation. Without bells and whistles, UniHead can unify these visual tasks via a single visual head design and achieve comparable performance compared to expert models developed for each task.We hope our simple and universal UniHead will serve as a solid baseline and help promote universal visual perception research. Code and models are available at https://github.com/Sense-X/UniHead.

Siyuan Yang, Lu Zhang, Liqian Ma et al.

Image inpainting aims to fill in the missing pixels with visually coherent and semantically plausible content. Despite the great progress brought from deep generative models, this task still suffers from i. the difficulties in large-scale realistic data collection and costly model training; and ii. the intrinsic limitations in the traditionally user-defined binary masks on objects with unclear boundaries or transparent texture. In this paper, we propose MagicRemover, a tuning-free method that leverages the powerful diffusion models for text-guided image inpainting. We introduce an attention guidance strategy to constrain the sampling process of diffusion models, enabling the erasing of instructed areas and the restoration of occluded content. We further propose a classifier optimization algorithm to facilitate the denoising stability within less sampling steps. Extensive comparisons are conducted among our MagicRemover and state-of-the-art methods including quantitative evaluation and user study, demonstrating the significant improvement of MagicRemover on high-quality image inpainting. We will release our code at https://github.com/exisas/Magicremover.

Manyuan Zhang, Guanglu Song, Yu Liu et al.

The introduction of DETR represents a new paradigm for object detection. However, its decoder conducts classification and box localization using shared queries and cross-attention layers, leading to suboptimal results. We observe that different regions of interest in the visual feature map are suitable for performing query classification and box localization tasks, even for the same object. Salient regions provide vital information for classification, while the boundaries around them are more favorable for box regression. Unfortunately, such spatial misalignment between these two tasks greatly hinders DETR's training. Therefore, in this work, we focus on decoupling localization and classification tasks in DETR. To achieve this, we introduce a new design scheme called spatially decoupled DETR (SD-DETR), which includes a task-aware query generation module and a disentangled feature learning process. We elaborately design the task-aware query initialization process and divide the cross-attention block in the decoder to allow the task-aware queries to match different visual regions. Meanwhile, we also observe that the prediction misalignment problem for high classification confidence and precise localization exists, so we propose an alignment loss to further guide the spatially decoupled DETR training. Through extensive experiments, we demonstrate that our approach achieves a significant improvement in MSCOCO datasets compared to previous work. For instance, we improve the performance of Conditional DETR by 4.5 AP. By spatially disentangling the two tasks, our method overcomes the misalignment problem and greatly improves the performance of DETR for object detection.

Yixin Guo, Yu Liu, Jianghao Li et al.

Zero-shot human-object interaction (HOI) detector is capable of generalizing to HOI categories even not encountered during training. Inspired by the impressive zero-shot capabilities offered by CLIP, latest methods strive to leverage CLIP embeddings for improving zero-shot HOI detection. However, these embedding-based methods train the classifier on seen classes only, inevitably resulting in seen-unseen confusion for the model during inference. Besides, we find that using prompt-tuning and adapters further increases the gap between seen and unseen accuracy. To tackle this challenge, we present the first generation-based model using CLIP for zero-shot HOI detection, coined HOIGen. It allows to unlock the potential of CLIP for feature generation instead of feature extraction only. To achieve it, we develop a CLIP-injected feature generator in accordance with the generation of human, object and union features. Then, we extract realistic features of seen samples and mix them with synthetic features together, allowing the model to train seen and unseen classes jointly. To enrich the HOI scores, we construct a generative prototype bank in a pairwise HOI recognition branch, and a multi-knowledge prototype bank in an image-wise HOI recognition branch, respectively. Extensive experiments on HICO-DET benchmark demonstrate our HOIGen achieves superior performance for both seen and unseen classes under various zero-shot settings, compared with other top-performing methods. Code is available at: https://github.com/soberguo/HOIGen

Xizhe Xue, Dongdong Yu, Lingqiao Liu et al.

Open-World Instance Segmentation (OWIS) is an emerging research topic that aims to segment class-agnostic object instances from images. The mainstream approaches use a two-stage segmentation framework, which first locates the candidate object bounding boxes and then performs instance segmentation. In this work, we instead promote a single-stage framework for OWIS. We argue that the end-to-end training process in the single-stage framework can be more convenient for directly regularizing the localization of class-agnostic object pixels. Based on the single-stage instance segmentation framework, we propose a regularization model to predict foreground pixels and use its relation to instance segmentation to construct a cross-task consistency loss. We show that such a consistency loss could alleviate the problem of incomplete instance annotation -- a common problem in the existing OWIS datasets. We also show that the proposed loss lends itself to an effective solution to semi-supervised OWIS that could be considered an extreme case that all object annotations are absent for some images. Our extensive experiments demonstrate that the proposed method achieves impressive results in both fully-supervised and semi-supervised settings. Compared to SOTA methods, the proposed method significantly improves the $AP_{100}$ score by 4.75\% in UVO$\rightarrow$UVO setting and 4.05\% in COCO$\rightarrow$UVO setting. In the case of semi-supervised learning, our model learned with only 30\% labeled data, even outperforms its fully-supervised counterpart with 50\% labeled data. The code will be released soon.

Cunjuan Zhu, Qi Jia, Wei Chen et al.

Video-Text Retrieval (VTR) aims to search for the most relevant video related to the semantics in a given sentence, and vice versa. In general, this retrieval task is composed of four successive steps: video and textual feature representation extraction, feature embedding and matching, and objective functions. In the last, a list of samples retrieved from the dataset is ranked based on their matching similarities to the query. In recent years, significant and flourishing progress has been achieved by deep learning techniques, however, VTR is still a challenging task due to the problems like how to learn an efficient spatial-temporal video feature and how to narrow the cross-modal gap. In this survey, we review and summarize over 100 research papers related to VTR, demonstrate state-of-the-art performance on several commonly benchmarked datasets, and discuss potential challenges and directions, with the expectation to provide some insights for researchers in the field of video-text retrieval.

Siyu Xing, Chen Gong, Hewei Guo et al.

Existing GAN inversion methods work brilliantly in reconstructing high-quality (HQ) images while struggling with more common low-quality (LQ) inputs in practical application. To address this issue, we propose Unsupervised Domain Adaptation (UDA) in the inversion process, namely UDA-inversion, for effective inversion and editing of both HQ and LQ images. Regarding unpaired HQ images as the source domain and LQ images as the unlabeled target domain, we introduce a theoretical guarantee: loss value in the target domain is upper-bounded by loss in the source domain and a novel discrepancy function measuring the difference between two domains. Following that, we can only minimize this upper bound to obtain accurate latent codes for HQ and LQ images. Thus, constructive representations of HQ images can be spontaneously learned and transformed into LQ images without supervision. UDA-Inversion achieves a better PSNR of 22.14 on FFHQ dataset and performs comparably to supervised methods.

Yueqin Yin, Lianghua Huang, Yu Liu et al.

Recent generative models show impressive results in photo-realistic image generation. However, artifacts often inevitably appear in the generated results, leading to downgraded user experience and reduced performance in downstream tasks. This work aims to develop a plugin post-processing module for diverse generative models, which can faithfully restore images from diverse generative artifacts. This is challenging because: (1) Unlike traditional degradation patterns, generative artifacts are non-linear and the transformation function is highly complex. (2) There are no readily available artifact-image pairs. (3) Different from model-specific anti-artifact methods, a model-agnostic framework views the generator as a black-box machine and has no access to the architecture details. In this work, we first design a group of mechanisms to simulate generative artifacts of popular generators (i.e., GANs, autoregressive models, and diffusion models), given real images. Second, we implement the model-agnostic anti-artifact framework as an image-to-image diffusion model, due to its advantage in generation quality and capacity. Finally, we design a conditioning scheme for the diffusion model to enable both blind and non-blind image restoration. A guidance parameter is also introduced to allow for a trade-off between restoration accuracy and image quality. Extensive experiments show that our method significantly outperforms previous approaches on the proposed datasets and real-world artifact images.

Bingqi Ma, Guanglu Song, Boxiao Liu et al.

Learning robust feature representation from large-scale noisy faces stands out as one of the key challenges in high-performance face recognition. Recent attempts have been made to cope with this challenge by alleviating the intra-class conflict and inter-class conflict. However, the unconstrained noise type in each conflict still makes it difficult for these algorithms to perform well. To better understand this, we reformulate the noise type of each class in a more fine-grained manner as N-identities|K^C-clusters. Different types of noisy faces can be generated by adjusting the values of \nkc. Based on this unified formulation, we found that the main barrier behind the noise-robust representation learning is the flexibility of the algorithm under different N, K, and C. For this potential problem, we propose a new method, named Evolving Sub-centers Learning~(ESL), to find optimal hyperplanes to accurately describe the latent space of massive noisy faces. More specifically, we initialize M sub-centers for each class and ESL encourages it to be automatically aligned to N-identities|K^C-clusters faces via producing, merging, and dropping operations. Images belonging to the same identity in noisy faces can effectively converge to the same sub-center and samples with different identities will be pushed away. We inspect its effectiveness with an elaborate ablation study on the synthetic noisy dataset with different N, K, and C. Without any bells and whistles, ESL can achieve significant performance gains over state-of-the-art methods on large-scale noisy faces

Rukai Wei, Heng Cui, Yu Liu et al.

Implementing cross-modal hashing between 2D images and 3D point-cloud data is a growing concern in real-world retrieval systems. Simply applying existing cross-modal approaches to this new task fails to adequately capture latent multi-modal semantics and effectively bridge the modality gap between 2D and 3D. To address these issues without relying on hand-crafted labels, we propose contrastive masked autoencoders based self-supervised hashing (CMAH) for retrieval between images and point-cloud data. We start by contrasting 2D-3D pairs and explicitly constraining them into a joint Hamming space. This contrastive learning process ensures robust discriminability for the generated hash codes and effectively reduces the modality gap. Moreover, we utilize multi-modal auto-encoders to enhance the model's understanding of multi-modal semantics. By completing the masked image/point-cloud data modeling task, the model is encouraged to capture more localized clues. In addition, the proposed multi-modal fusion block facilitates fine-grained interactions among different modalities. Extensive experiments on three public datasets demonstrate that the proposed CMAH significantly outperforms all baseline methods.

Lu Han, Yu Liu, Qiwen Deng et al.

Time Series Foundation Models (TSFMs) have achieved remarkable success through large-scale pretraining. However, their design primarily targets real-valued series, limiting their ability to handle general forecasting tasks involving diverse and often heterogeneous covariates--such as categorical variables and multimodal data (e.g., images, text)--which are typically task-specific and difficult to leverage during pretraining. To address this gap, we propose Unified Covariate Adaptation (UniCA), a framework to bridge TSFMs with general covariate-aware forecasting. UniCA first performs covariate homogenization to transform heterogeneous covariates into high-level homogeneous series representations and then fuses them via a unified attention-based fusion mechanism. UniCA is compatible and universal for adaptation with both homogeneous and heterogeneous covariates, incorporating extra covariate information while preserving the generalization ability of TSFMs.Extensive experiments on multiple unimodal and multimodal covariate-aware forecasting benchmarks demonstrate the superiority of UniCA, highlighting the promise of covariate-aware TSFM adaptation in real-world forecasting scenarios. Codes are released on https://github.com/hanlu-nju/UniCA.

Weimin Wang, Yingxu Deng, Zezeng Li et al.

This paper introduces a novel method for reconstructing meshes from sparse point clouds by predicting edge connection. Existing implicit methods usually produce superior smooth and watertight meshes due to the isosurface extraction algorithms~(e.g., Marching Cubes). However, these methods become memory and computationally intensive with increasing resolution. Explicit methods are more efficient by directly forming the face from points. Nevertheless, the challenge of selecting appropriate faces from enormous candidates often leads to undesirable faces and holes. Moreover, the reconstruction performance of both approaches tends to degrade when the point cloud gets sparse. To this end, we propose MEsh Reconstruction via edGE~(MergeNet), which converts mesh reconstruction into local connectivity prediction problems. Specifically, MergeNet learns to extract the features of candidate edges and regress their distances to the underlying surface. Consequently, the predicted distance is utilized to filter out edges that lay on surfaces. Finally, the meshes are reconstructed by refining the triangulations formed by these edges. Extensive experiments on synthetic and real-scanned datasets demonstrate the superiority of MergeNet to SoTA explicit methods.

Zhuofan Zong, Kunchang Li, Guanglu Song et al.

Vision transformers (ViTs) have become the popular structures and outperformed convolutional neural networks (CNNs) on various vision tasks. However, such powerful transformers bring a huge computation burden, because of the exhausting token-to-token comparison. The previous works focus on dropping insignificant tokens to reduce the computational cost of ViTs. But when the dropping ratio increases, this hard manner will inevitably discard the vital tokens, which limits its efficiency. To solve the issue, we propose a generic self-slimmed learning approach for vanilla ViTs, namely SiT. Specifically, we first design a novel Token Slimming Module (TSM), which can boost the inference efficiency of ViTs by dynamic token aggregation. As a general method of token hard dropping, our TSM softly integrates redundant tokens into fewer informative ones. It can dynamically zoom visual attention without cutting off discriminative token relations in the images, even with a high slimming ratio. Furthermore, we introduce a concise Feature Recalibration Distillation (FRD) framework, wherein we design a reverse version of TSM (RTSM) to recalibrate the unstructured token in a flexible auto-encoder manner. Due to the similar structure between teacher and student, our FRD can effectively leverage structure knowledge for better convergence. Finally, we conduct extensive experiments to evaluate our SiT. It demonstrates that our method can speed up ViTs by 1.7x with negligible accuracy drop, and even speed up ViTs by 3.6x while maintaining 97% of their performance. Surprisingly, by simply arming LV-ViT with our SiT, we achieve new state-of-the-art performance on ImageNet. Code is available at https://github.com/Sense-X/SiT.

Haokui Zhang, Chunhua Shen, Ying Li et al.

Accuracy of depth estimation from static images has been significantly improved recently, by exploiting hierarchical features from deep convolutional neural networks (CNNs). Compared with static images, vast information exists among video frames and can be exploited to improve the depth estimation performance. In this work, we focus on exploring temporal information from monocular videos for depth estimation. Specifically, we take the advantage of convolutional long short-term memory (CLSTM) and propose a novel spatial-temporal CSLTM (ST-CLSTM) structure. Our ST-CLSTM structure can capture not only the spatial features but also the temporal correlations/consistency among consecutive video frames with negligible increase in computational cost. Additionally, in order to maintain the temporal consistency among the estimated depth frames, we apply the generative adversarial learning scheme and design a temporal consistency loss. The temporal consistency loss is combined with the spatial loss to update the model in an end-to-end fashion. By taking advantage of the temporal information, we build a video depth estimation framework that runs in real-time and generates visually pleasant results. Moreover, our approach is flexible and can be generalized to most existing depth estimation frameworks. Code is available at: https://tinyurl.com/STCLSTM

Yu Liu, Hongyang Li, Junjie Yan et al.

Since convolutional neural network (CNN) lacks an inherent mechanism to handle large scale variations, we always need to compute feature maps multiple times for multi-scale object detection, which has the bottleneck of computational cost in practice. To address this, we devise a recurrent scale approximation (RSA) to compute feature map once only, and only through this map can we approximate the rest maps on other levels. At the core of RSA is the recursive rolling out mechanism: given an initial map at a particular scale, it generates the prediction at a smaller scale that is half the size of input. To further increase efficiency and accuracy, we (a): design a scale-forecast network to globally predict potential scales in the image since there is no need to compute maps on all levels of the pyramid. (b): propose a landmark retracing network (LRN) to trace back locations of the regressed landmarks and generate a confidence score for each landmark; LRN can effectively alleviate false positives caused by the accumulated error in RSA. The whole system can be trained end-to-end in a unified CNN framework. Experiments demonstrate that our proposed algorithm is superior against state-of-the-art methods on face detection benchmarks and achieves comparable results for generic proposal generation. The source code of RSA is available at github.com/sciencefans/RSA-for-object-detection.

Xingyu Zeng, Wanli Ouyang, Junjie Yan et al.

The visual cues from multiple support regions of different sizes and resolutions are complementary in classifying a candidate box in object detection. Effective integration of local and contextual visual cues from these regions has become a fundamental problem in object detection. In this paper, we propose a gated bi-directional CNN (GBD-Net) to pass messages among features from different support regions during both feature learning and feature extraction. Such message passing can be implemented through convolution between neighboring support regions in two directions and can be conducted in various layers. Therefore, local and contextual visual patterns can validate the existence of each other by learning their nonlinear relationships and their close interactions are modeled in a more complex way. It is also shown that message passing is not always helpful but dependent on individual samples. Gated functions are therefore needed to control message transmission, whose on-or-offs are controlled by extra visual evidence from the input sample. The effectiveness of GBD-Net is shown through experiments on three object detection datasets, ImageNet, Pascal VOC2007 and Microsoft COCO. This paper also shows the details of our approach in wining the ImageNet object detection challenge of 2016, with source code provided on \url{https://github.com/craftGBD/craftGBD}.

Dongzhi Jiang, Guanglu Song, Xiaoshi Wu et al.

Diffusion models have demonstrated great success in the field of text-to-image generation. However, alleviating the misalignment between the text prompts and images is still challenging. The root reason behind the misalignment has not been extensively investigated. We observe that the misalignment is caused by inadequate token attention activation. We further attribute this phenomenon to the diffusion model's insufficient condition utilization, which is caused by its training paradigm. To address the issue, we propose CoMat, an end-to-end diffusion model fine-tuning strategy with an image-to-text concept matching mechanism. We leverage an image captioning model to measure image-to-text alignment and guide the diffusion model to revisit ignored tokens. A novel attribute concentration module is also proposed to address the attribute binding problem. Without any image or human preference data, we use only 20K text prompts to fine-tune SDXL to obtain CoMat-SDXL. Extensive experiments show that CoMat-SDXL significantly outperforms the baseline model SDXL in two text-to-image alignment benchmarks and achieves start-of-the-art performance.

Zhuofan Zong, Dongzhi Jiang, Bingqi Ma et al. · tsinghua

Significant achievements in personalization of diffusion models have been witnessed. Conventional tuning-free methods mostly encode multiple reference images by averaging their image embeddings as the injection condition, but such an image-independent operation cannot perform interaction among images to capture consistent visual elements within multiple references. Although the tuning-based Low-Rank Adaptation (LoRA) can effectively extract consistent elements within multiple images through the training process, it necessitates specific finetuning for each distinct image group. This paper introduces EasyRef, a novel plug-and-play adaptation method that enables diffusion models to be conditioned on multiple reference images and the text prompt. To effectively exploit consistent visual elements within multiple images, we leverage the multi-image comprehension and instruction-following capabilities of the multimodal large language model (MLLM), prompting it to capture consistent visual elements based on the instruction. Besides, injecting the MLLM's representations into the diffusion process through adapters can easily generalize to unseen domains, mining the consistent visual elements within unseen data. To mitigate computational costs and enhance fine-grained detail preservation, we introduce an efficient reference aggregation strategy and a progressive training scheme. Finally, we introduce MRBench, a new multi-reference image generation benchmark. Experimental results demonstrate EasyRef surpasses both tuning-free methods like IP-Adapter and tuning-based methods like LoRA, achieving superior aesthetic quality and robust zero-shot generalization across diverse domains.

Qinying Liu, Wei Wu, Kecheng Zheng et al.

The crux of learning vision-language models is to extract semantically aligned information from visual and linguistic data. Existing attempts usually face the problem of coarse alignment, e.g., the vision encoder struggles in localizing an attribute-specified object. In this work, we propose an embarrassingly simple approach to better align image and text features with no need of additional data formats other than image-text pairs. Concretely, given an image and its paired text, we manage to parse objects (e.g., cat) and attributes (e.g., black) from the description, which are highly likely to exist in the image. It is noteworthy that the parsing pipeline is fully automatic and thus enjoys good scalability. With these parsed semantics as supervision signals, we can complement the commonly used image-text contrastive loss with the multi-tag classification loss. Extensive experimental results on a broad suite of semantic segmentation datasets substantiate the average 5.2\% improvement of our framework over existing alternatives. Furthermore, the visualization results indicate that attribute supervision makes vision-language models accurately localize attribute-specified objects. Project page can be found at https://qinying-liu.github.io/Tag-Align.

Yiyang Wang, Xi Chen, Xiaogang Xu et al.

In spite of recent progress, image diffusion models still produce artifacts. A common solution is to leverage the feedback provided by quality assessment systems or human annotators to optimize the model, where images are generally rated in their entirety. In this work, we believe problem-solving starts with identification, yielding the request that the model should be aware of not just the presence of defects in an image, but their specific locations. Motivated by this, we propose DiffDoctor, a two-stage pipeline to assist image diffusion models in generating fewer artifacts. Concretely, the first stage targets developing a robust artifact detector, for which we collect a dataset of over 1M flawed synthesized images and set up an efficient human-in-the-loop annotation process, incorporating a carefully designed class-balance strategy. The learned artifact detector is then involved in the second stage to optimize the diffusion model by providing pixel-level feedback. Extensive experiments on text-to-image diffusion models demonstrate the effectiveness of our artifact detector as well as the soundness of our diagnose-then-treat design.

Jian Peng, Dingqi Ye, Bo Tang et al.

Learning from a sequence of tasks for a lifetime is essential for an agent towards artificial general intelligence. This requires the agent to continuously learn and memorize new knowledge without interference. This paper first demonstrates a fundamental issue of lifelong learning using neural networks, named anterograde forgetting, i.e., preserving and transferring memory may inhibit the learning of new knowledge. This is attributed to the fact that the learning capacity of a neural network will be reduced as it keeps memorizing historical knowledge, and the fact that conceptual confusion may occur as it transfers irrelevant old knowledge to the current task. This work proposes a general framework named Cycled Memory Networks (CMN) to address the anterograde forgetting in neural networks for lifelong learning. The CMN consists of two individual memory networks to store short-term and long-term memories to avoid capacity shrinkage. A transfer cell is designed to connect these two memory networks, enabling knowledge transfer from the long-term memory network to the short-term memory network to mitigate the conceptual confusion, and a memory consolidation mechanism is developed to integrate short-term knowledge into the long-term memory network for knowledge accumulation. Experimental results demonstrate that the CMN can effectively address the anterograde forgetting on several task-related, task-conflict, class-incremental and cross-domain benchmarks.

Yifan Chang, Wenbo Li, Jian Peng et al.

Humans' continual learning (CL) ability is closely related to Stability Versus Plasticity Dilemma that describes how humans achieve ongoing learning capacity and preservation for learned information. The notion of CL has always been present in artificial intelligence (AI) since its births. This paper proposes a comprehensive review of CL. Different from previous reviews that mainly focus on the catastrophic forgetting phenomenon in CL, this paper surveys CL from a more macroscopic perspective based on the Stability Versus Plasticity mechanism. Analogous to biological counterpart, "smart" AI agents are supposed to i) remember previously learned information (information retrospection); ii) infer on new information continuously (information prospection:); iii) transfer useful information (information transfer), to achieve high-level CL. According to the taxonomy, evaluation metrics, algorithms, applications as well as some open issues are then introduced. Our main contributions concern i) rechecking CL from the level of artificial general intelligence; ii) providing a detailed and extensive overview on CL topics; iii) presenting some novel ideas on the potential development of CL.

Jing Shao, Siyu Chen, Yangguang Li et al.

Enormous waves of technological innovations over the past several years, marked by the advances in AI technologies, are profoundly reshaping the industry and the society. However, down the road, a key challenge awaits us, that is, our capability of meeting rapidly-growing scenario-specific demands is severely limited by the cost of acquiring a commensurate amount of training data. This difficult situation is in essence due to limitations of the mainstream learning paradigm: we need to train a new model for each new scenario, based on a large quantity of well-annotated data and commonly from scratch. In tackling this fundamental problem, we move beyond and develop a new learning paradigm named INTERN. By learning with supervisory signals from multiple sources in multiple stages, the model being trained will develop strong generalizability. We evaluate our model on 26 well-known datasets that cover four categories of tasks in computer vision. In most cases, our models, adapted with only 10% of the training data in the target domain, outperform the counterparts trained with the full set of data, often by a significant margin. This is an important step towards a promising prospect where such a model with general vision capability can dramatically reduce our reliance on data, thus expediting the adoption of AI technologies. Furthermore, revolving around our new paradigm, we also introduce a new data system, a new architecture, and a new benchmark, which, together, form a general vision ecosystem to support its future development in an open and inclusive manner. See project website at https://opengvlab.shlab.org.cn .

Xiaoyu Chen, Chen Gong, Qiang He et al.

Variational autoencoders (VAEs), as an important aspect of generative models, have received a lot of research interests and reached many successful applications. However, it is always a challenge to achieve the consistency between the learned latent distribution and the prior latent distribution when optimizing the evidence lower bound (ELBO), and finally leads to an unsatisfactory performance in data generation. In this paper, we propose a latent distribution consistency approach to avoid such substantial inconsistency between the posterior and prior latent distributions in ELBO optimizing. We name our method as latent distribution consistency VAE (LDC-VAE). We achieve this purpose by assuming the real posterior distribution in latent space as a Gibbs form, and approximating it by using our encoder. However, there is no analytical solution for such Gibbs posterior in approximation, and traditional approximation ways are time consuming, such as using the iterative sampling-based MCMC. To address this problem, we use the Stein Variational Gradient Descent (SVGD) to approximate the Gibbs posterior. Meanwhile, we use the SVGD to train a sampler net which can obtain efficient samples from the Gibbs posterior. Comparative studies on the popular image generation datasets show that our method has achieved comparable or even better performance than several powerful improvements of VAEs.

Wei Chen, Yu Liu, Weiping Wang et al.

In recent years a vast amount of visual content has been generated and shared from many fields, such as social media platforms, medical imaging, and robotics. This abundance of content creation and sharing has introduced new challenges, particularly that of searching databases for similar content-Content Based Image Retrieval (CBIR)-a long-established research area in which improved efficiency and accuracy are needed for real-time retrieval. Artificial intelligence has made progress in CBIR and has significantly facilitated the process of instance search. In this survey we review recent instance retrieval works that are developed based on deep learning algorithms and techniques, with the survey organized by deep network architecture types, deep features, feature embedding and aggregation methods, and network fine-tuning strategies. Our survey considers a wide variety of recent methods, whereby we identify milestone work, reveal connections among various methods and present the commonly used benchmarks, evaluation results, common challenges, and propose promising future directions.

Wei Chen, Yu Liu, Weiping Wang et al.

In this paper, we consider the problem of fine-grained image retrieval in an incremental setting, when new categories are added over time. On the one hand, repeatedly training the representation on the extended dataset is time-consuming. On the other hand, fine-tuning the learned representation only with the new classes leads to catastrophic forgetting. To this end, we propose an incremental learning method to mitigate retrieval performance degradation caused by the forgetting issue. Without accessing any samples of the original classes, the classifier of the original network provides soft "labels" to transfer knowledge to train the adaptive network, so as to preserve the previous capability for classification. More importantly, a regularization function based on Maximum Mean Discrepancy is devised to minimize the discrepancy of new classes features from the original network and the adaptive network, respectively. Extensive experiments on two datasets show that our method effectively mitigates the catastrophic forgetting on the original classes while achieving high performance on the new classes.

Thierry Deruyttere, Simon Vandenhende, Dusan Grujicic et al.

The task of visual grounding requires locating the most relevant region or object in an image, given a natural language query. So far, progress on this task was mostly measured on curated datasets, which are not always representative of human spoken language. In this work, we deviate from recent, popular task settings and consider the problem under an autonomous vehicle scenario. In particular, we consider a situation where passengers can give free-form natural language commands to a vehicle which can be associated with an object in the street scene. To stimulate research on this topic, we have organized the \emph{Commands for Autonomous Vehicles} (C4AV) challenge based on the recent \emph{Talk2Car} dataset (URL: https://www.aicrowd.com/challenges/eccv-2020-commands-4-autonomous-vehicles). This paper presents the results of the challenge. First, we compare the used benchmark against existing datasets for visual grounding. Second, we identify the aspects that render top-performing models successful, and relate them to existing state-of-the-art models for visual grounding, in addition to detecting potential failure cases by evaluating on carefully selected subsets. Finally, we discuss several possibilities for future work.

Ling Yang, Liangliang Li, Zilun Zhang et al.

Most graph-network-based meta-learning approaches model instance-level relation of examples. We extend this idea further to explicitly model the distribution-level relation of one example to all other examples in a 1-vs-N manner. We propose a novel approach named distribution propagation graph network (DPGN) for few-shot learning. It conveys both the distribution-level relations and instance-level relations in each few-shot learning task. To combine the distribution-level relations and instance-level relations for all examples, we construct a dual complete graph network which consists of a point graph and a distribution graph with each node standing for an example. Equipped with dual graph architecture, DPGN propagates label information from labeled examples to unlabeled examples within several update generations. In extensive experiments on few-shot learning benchmarks, DPGN outperforms state-of-the-art results by a large margin in 5% $\sim$ 12% under supervised setting and 7% $\sim$ 13% under semi-supervised setting. Code will be released.

Yu Liu, Guanglu Song, Yuhang Zang et al.

This article introduces the solutions of the two champion teams, `MMfruit' for the detection track and `MMfruitSeg' for the segmentation track, in OpenImage Challenge 2019. It is commonly known that for an object detector, the shared feature at the end of the backbone is not appropriate for both classification and regression, which greatly limits the performance of both single stage detector and Faster RCNN \cite{ren2015faster} based detector. In this competition, we observe that even with a shared feature, different locations in one object has completely inconsistent performances for the two tasks. \textit{E.g. the features of salient locations are usually good for classification, while those around the object edge are good for regression.} Inspired by this, we propose the Decoupling Head (DH) to disentangle the object classification and regression via the self-learned optimal feature extraction, which leads to a great improvement. Furthermore, we adjust the soft-NMS algorithm to adj-NMS to obtain stable performance improvement. Finally, a well-designed ensemble strategy via voting the bounding box location and confidence is proposed. We will also introduce several training/inferencing strategies and a bag of tricks that give minor improvement. Given those masses of details, we train and aggregate 28 global models with various backbones, heads and 3+2 expert models, and achieves the 1st place on the OpenImage 2019 Object Detection Challenge on the both public and private leadboards. Given such good instance bounding box, we further design a simple instance-level semantic segmentation pipeline and achieve the 1st place on the segmentation challenge.

Guanglu Song, Yu Liu, Xiaogang Wang

The ``shared head for classification and localization'' (sibling head), firstly denominated in Fast RCNN~\cite{girshick2015fast}, has been leading the fashion of the object detection community in the past five years. This paper provides the observation that the spatial misalignment between the two object functions in the sibling head can considerably hurt the training process, but this misalignment can be resolved by a very simple operator called task-aware spatial disentanglement (TSD). Considering the classification and regression, TSD decouples them from the spatial dimension by generating two disentangled proposals for them, which are estimated by the shared proposal. This is inspired by the natural insight that for one instance, the features in some salient area may have rich information for classification while these around the boundary may be good at bounding box regression. Surprisingly, this simple design can boost all backbones and models on both MS COCO and Google OpenImage consistently by ~3% mAP. Further, we propose a progressive constraint to enlarge the performance margin between the disentangled and the shared proposals, and gain ~1% more mAP. We show the \algname{} breaks through the upper bound of nowadays single-model detector by a large margin (mAP 49.4 with ResNet-101, 51.2 with SENet154), and is the core model of our 1st place solution on the Google OpenImage Challenge 2019.

Yu Liu, Jie Li, Qingsen Yan et al.

This paper tackles the problem of data fusion in the semantic scene completion (SSC) task, which can simultaneously deal with semantic labeling and scene completion. RGB images contain texture details of the object(s) which are vital for semantic scene understanding. Meanwhile, depth images capture geometric clues of high relevance for shape completion. Using both RGB and depth images can further boost the accuracy of SSC over employing one modality in isolation. We propose a 3D gated recurrent fusion network (GRFNet), which learns to adaptively select and fuse the relevant information from depth and RGB by making use of the gate and memory modules. Based on the single-stage fusion, we further propose a multi-stage fusion strategy, which could model the correlations among different stages within the network. Extensive experiments on two benchmark datasets demonstrate the superior performance and the effectiveness of the proposed GRFNet for data fusion in SSC. Code will be made available.

Yu Liu, Xuhui Jia, Mingxing Tan et al.

Standard Knowledge Distillation (KD) approaches distill the knowledge of a cumbersome teacher model into the parameters of a student model with a pre-defined architecture. However, the knowledge of a neural network, which is represented by the network's output distribution conditioned on its input, depends not only on its parameters but also on its architecture. Hence, a more generalized approach for KD is to distill the teacher's knowledge into both the parameters and architecture of the student. To achieve this, we present a new Architecture-aware Knowledge Distillation (AKD) approach that finds student models (pearls for the teacher) that are best for distilling the given teacher model. In particular, we leverage Neural Architecture Search (NAS), equipped with our KD-guided reward, to search for the best student architectures for a given teacher. Experimental results show our proposed AKD consistently outperforms the conventional NAS plus KD approach, and achieves state-of-the-art results on the ImageNet classification task under various latency settings. Furthermore, the best AKD student architecture for the ImageNet classification task also transfers well to other tasks such as million level face recognition and ensemble learning.

Jinwei Zhao, Qizhou Wang, Fuqiang Zhang et al.

Real artificial intelligence always has been focused on by many machine learning researchers, especially in the area of deep learning. However deep neural network is hard to be understood and explained, and sometimes, even metaphysics. The reason is, we believe that: the network is essentially a perceptual model. Therefore, we believe that in order to complete complex intelligent activities from simple perception, it is necessary to con-struct another interpretable logical network to form accurate and reasonable responses and explanations to external things. Researchers like Bolei Zhou and Quanshi Zhang have found many explanatory rules for deep feature extraction aimed at the feature extraction stage of convolution neural network. However, although researchers like Marco Gori have also made great efforts to improve the interpretability of the fully connected layers of the network, the problem is also very difficult. This paper firstly analyzes its reason. Then a method of constructing logical network based on the fully connected layers and extracting logical relation between input and output of the layers is proposed. The game process between perceptual learning and logical abstract cognitive learning is implemented to improve the interpretable performance of deep learning process and deep learning model. The benefits of our approach are illustrated on benchmark data sets and in real-world experiments.

Yu Liu, Lingqiao Liu, Haokui Zhang et al.

This paper tackles the problem of video object segmentation. We are specifically concerned with the task of segmenting all pixels of a target object in all frames, given the annotation mask in the first frame. Even when such annotation is available this remains a challenging problem because of the changing appearance and shape of the object over time. In this paper, we tackle this task by formulating it as a meta-learning problem, where the base learner grasping the semantic scene understanding for a general type of objects, and the meta learner quickly adapting the appearance of the target object with a few examples. Our proposed meta-learning method uses a closed form optimizer, the so-called "ridge regression", which has been shown to be conducive for fast and better training convergence. Moreover, we propose a mechanism, named "block splitting", to further speed up the training process as well as to reduce the number of learning parameters. In comparison with the-state-of-the art methods, our proposed framework achieves significant boost up in processing speed, while having very competitive performance compared to the best performing methods on the widely used datasets.

Baoyun Peng, Xiao Jin, Jiaheng Liu et al.

Most teacher-student frameworks based on knowledge distillation (KD) depend on a strong congruent constraint on instance level. However, they usually ignore the correlation between multiple instances, which is also valuable for knowledge transfer. In this work, we propose a new framework named correlation congruence for knowledge distillation (CCKD), which transfers not only the instance-level information, but also the correlation between instances. Furthermore, a generalized kernel method based on Taylor series expansion is proposed to better capture the correlation between instances. Empirical experiments and ablation studies on image classification tasks (including CIFAR-100, ImageNet-1K) and metric learning tasks (including ReID and Face Recognition) show that the proposed CCKD substantially outperforms the original KD and achieves state-of-the-art accuracy compared with other SOTA KD-based methods. The CCKD can be easily deployed in the majority of the teacher-student framework such as KD and hint-based learning methods.

Rui Liu, Yu Liu, Xinyu Gong et al.

Flow-based generative models show great potential in image synthesis due to its reversible pipeline and exact log-likelihood target, yet it suffers from weak ability for conditional image synthesis, especially for multi-label or unaware conditions. This is because the potential distribution of image conditions is hard to measure precisely from its latent variable $z$. In this paper, based on modeling a joint probabilistic density of an image and its conditions, we propose a novel flow-based generative model named conditional adversarial generative flow (CAGlow). Instead of disentangling attributes from latent space, we blaze a new trail for learning an encoder to estimate the mapping from condition space to latent space in an adversarial manner. Given a specific condition $c$, CAGlow can encode it to a sampled $z$, and then enable robust conditional image synthesis in complex situations like combining person identity with multiple attributes. The proposed CAGlow can be implemented in both supervised and unsupervised manners, thus can synthesize images with conditional information like categories, attributes, and even some unknown properties. Extensive experiments show that CAGlow ensures the independence of different conditions and outperforms regular Glow to a significant extent.

Jie Li, Yu Liu, Dong Gong et al.

RGB images differentiate from depth images as they carry more details about the color and texture information, which can be utilized as a vital complementary to depth for boosting the performance of 3D semantic scene completion (SSC). SSC is composed of 3D shape completion (SC) and semantic scene labeling while most of the existing methods use depth as the sole input which causes the performance bottleneck. Moreover, the state-of-the-art methods employ 3D CNNs which have cumbersome networks and tremendous parameters. We introduce a light-weight Dimensional Decomposition Residual network (DDR) for 3D dense prediction tasks. The novel factorized convolution layer is effective for reducing the network parameters, and the proposed multi-scale fusion mechanism for depth and color image can improve the completion and segmentation accuracy simultaneously. Our method demonstrates excellent performance on two public datasets. Compared with the latest method SSCNet, we achieve 5.9% gains in SC-IoU and 5.7% gains in SSC-IOU, albeit with only 21% network parameters and 16.6% FLOPs employed compared with that of SSCNet.

Yu Liu, Lingqiao Liu, Hamid Rezatofighi et al.

As the post-processing step for object detection, non-maximum suppression (GreedyNMS) is widely used in most of the detectors for many years. It is efficient and accurate for sparse scenes, but suffers an inevitable trade-off between precision and recall in crowded scenes. To overcome this drawback, we propose a Pairwise-NMS to cure GreedyNMS. Specifically, a pairwise-relationship network that is based on deep learning is learned to predict if two overlapping proposal boxes contain two objects or zero/one object, which can handle multiple overlapping objects effectively. Through neatly coupling with GreedyNMS without losing efficiency, consistent improvements have been achieved in heavily occluded datasets including MOT15, TUD-Crossing and PETS. In addition, Pairwise-NMS can be integrated into any learning based detectors (Both of Faster-RCNN and DPM detectors are tested in this paper), thus building a bridge between GreedyNMS and end-to-end learning detectors.

Hang Zhou, Yu Liu, Ziwei Liu et al.

Talking face generation aims to synthesize a sequence of face images that correspond to a clip of speech. This is a challenging task because face appearance variation and semantics of speech are coupled together in the subtle movements of the talking face regions. Existing works either construct specific face appearance model on specific subjects or model the transformation between lip motion and speech. In this work, we integrate both aspects and enable arbitrary-subject talking face generation by learning disentangled audio-visual representation. We find that the talking face sequence is actually a composition of both subject-related information and speech-related information. These two spaces are then explicitly disentangled through a novel associative-and-adversarial training process. This disentangled representation has an advantage where both audio and video can serve as inputs for generation. Extensive experiments show that the proposed approach generates realistic talking face sequences on arbitrary subjects with much clearer lip motion patterns than previous work. We also demonstrate the learned audio-visual representation is extremely useful for the tasks of automatic lip reading and audio-video retrieval.

Guanglu Song, Yu Liu, Ming Jiang et al.

Fully convolutional neural network (FCN) has been dominating the game of face detection task for a few years with its congenital capability of sliding-window-searching with shared kernels, which boiled down all the redundant calculation, and most recent state-of-the-art methods such as Faster-RCNN, SSD, YOLO and FPN use FCN as their backbone. So here comes one question: Can we find a universal strategy to further accelerate FCN with higher accuracy, so could accelerate all the recent FCN-based methods? To analyze this, we decompose the face searching space into two orthogonal directions, `scale' and `spatial'. Only a few coordinates in the space expanded by the two base vectors indicate foreground. So if FCN could ignore most of the other points, the searching space and false alarm should be significantly boiled down. Based on this philosophy, a novel method named scale estimation and spatial attention proposal ($S^2AP$) is proposed to pay attention to some specific scales and valid locations in the image pyramid. Furthermore, we adopt a masked-convolution operation based on the attention result to accelerate FCN calculation. Experiments show that FCN-based method RPN can be accelerated by about $4\times$ with the help of $S^2AP$ and masked-FCN and at the same time it can also achieve the state-of-the-art on FDDB, AFW and MALF face detection benchmarks as well.

Yu Liu, Fangyin Wei, Jing Shao et al.

This paper proposes learning disentangled but complementary face features with minimal supervision by face identification. Specifically, we construct an identity Distilling and Dispelling Autoencoder (D2AE) framework that adversarially learns the identity-distilled features for identity verification and the identity-dispelled features to fool the verification system. Thanks to the design of two-stream cues, the learned disentangled features represent not only the identity or attribute but the complete input image. Comprehensive evaluations further demonstrate that the proposed features not only maintain state-of-the-art identity verification performance on LFW, but also acquire competitive discriminative power for face attribute recognition on CelebA and LFWA. Moreover, the proposed system is ready to semantically control the face generation/editing based on various identities and attributes in an unsupervised manner.

Huaxin Xiao, Yunchao Wei, Yu Liu et al.

The performance of deep learning based semantic segmentation models heavily depends on sufficient data with careful annotations. However, even the largest public datasets only provide samples with pixel-level annotations for rather limited semantic categories. Such data scarcity critically limits scalability and applicability of semantic segmentation models in real applications. In this paper, we propose a novel transferable semi-supervised semantic segmentation model that can transfer the learned segmentation knowledge from a few strong categories with pixel-level annotations to unseen weak categories with only image-level annotations, significantly broadening the applicable territory of deep segmentation models. In particular, the proposed model consists of two complementary and learnable components: a Label transfer Network (L-Net) and a Prediction transfer Network (P-Net). The L-Net learns to transfer the segmentation knowledge from strong categories to the images in the weak categories and produces coarse pixel-level semantic maps, by effectively exploiting the similar appearance shared across categories. Meanwhile, the P-Net tailors the transferred knowledge through a carefully designed adversarial learning strategy and produces refined segmentation results with better details. Integrating the L-Net and P-Net achieves 96.5% and 89.4% performance of the fully-supervised baseline using 50% and 0% categories with pixel-level annotations respectively on PASCAL VOC 2012. With such a novel transfer mechanism, our proposed model is easily generalizable to a variety of new categories, only requiring image-level annotations, and offers appealing scalability in real applications.

Hongyang Li, Yu Liu, Wanli Ouyang et al.

In this paper, we propose a zoom-out-and-in network for generating object proposals. A key observation is that it is difficult to classify anchors of different sizes with the same set of features. Anchors of different sizes should be placed accordingly based on different depth within a network: smaller boxes on high-resolution layers with a smaller stride while larger boxes on low-resolution counterparts with a larger stride. Inspired by the conv/deconv structure, we fully leverage the low-level local details and high-level regional semantics from two feature map streams, which are complimentary to each other, to identify the objectness in an image. A map attention decision (MAD) unit is further proposed to aggressively search for neuron activations among two streams and attend the most contributive ones on the feature learning of the final loss. The unit serves as a decisionmaker to adaptively activate maps along certain channels with the solely purpose of optimizing the overall training loss. One advantage of MAD is that the learned weights enforced on each feature channel is predicted on-the-fly based on the input context, which is more suitable than the fixed enforcement of a convolutional kernel. Experimental results on three datasets, including PASCAL VOC 2007, ImageNet DET, MS COCO, demonstrate the effectiveness of our proposed algorithm over other state-of-the-arts, in terms of average recall (AR) for region proposal and average precision (AP) for object detection.

Yu Liu, Jianshu Chen, Li Deng

We consider learning a sequence classifier without labeled data by using sequential output statistics. The problem is highly valuable since obtaining labels in training data is often costly, while the sequential output statistics (e.g., language models) could be obtained independently of input data and thus with low or no cost. To address the problem, we propose an unsupervised learning cost function and study its properties. We show that, compared to earlier works, it is less inclined to be stuck in trivial solutions and avoids the need for a strong generative model. Although it is harder to optimize in its functional form, a stochastic primal-dual gradient method is developed to effectively solve the problem. Experiment results on real-world datasets demonstrate that the new unsupervised learning method gives drastically lower errors than other baseline methods. Specifically, it reaches test errors about twice of those obtained by fully supervised learning.