Yihui He

Cindy Le, Congrui Hetang, Chendi Lin et al.

This paper presents a novel method to generate textures for 3D models given text prompts and 3D meshes. Additional depth information is taken into account to perform the Score Distillation Sampling (SDS) process with depth conditional Stable Diffusion. We ran our model over the open-source dataset Objaverse and conducted a user study to compare the results with those of various 3D texturing methods. We have shown that our model can generate more satisfactory results and produce various art styles for the same object. In addition, we achieved faster time when generating textures of comparable quality. We also conduct thorough ablation studies of how different factors may affect generation quality, including sampling steps, guidance scale, negative prompts, data augmentation, elevation range, and alternatives to SDS.

Congrui Hetang, Haoru Xue, Cindy Le et al.

We propose SAM-Road, an adaptation of the Segment Anything Model (SAM) for extracting large-scale, vectorized road network graphs from satellite imagery. To predict graph geometry, we formulate it as a dense semantic segmentation task, leveraging the inherent strengths of SAM. The image encoder of SAM is fine-tuned to produce probability masks for roads and intersections, from which the graph vertices are extracted via simple non-maximum suppression. To predict graph topology, we designed a lightweight transformer-based graph neural network, which leverages the SAM image embeddings to estimate the edge existence probabilities between vertices. Our approach directly predicts the graph vertices and edges for large regions without expensive and complex post-processing heuristics, and is capable of building complete road network graphs spanning multiple square kilometers in a matter of seconds. With its simple, straightforward, and minimalist design, SAM-Road achieves comparable accuracy with the state-of-the-art method RNGDet++, while being 40 times faster on the City-scale dataset. We thus demonstrate the power of a foundational vision model when applied to a graph learning task. The code is available at https://github.com/htcr/sam_road.

Yihui He

In this paper, we introduce a new channel pruning method to accelerate very deep convolutional neural networks. Given a trained CNN model, we propose an iterative two-step algorithm to effectively prune each layer, by a LASSO regression based channel selection and least square reconstruction. We further generalize this algorithm to multi-layer and multi-branch cases. Our method reduces the accumulated error and enhances the compatibility with various architectures. Our pruned VGG-16 achieves the state-of-the-art results by 5x speed-up along with only 0.3% increase of error. More importantly, our method is able to accelerate modern networks like ResNet, Xception and suffers only 1.4%, 1.0% accuracy loss under 2x speed-up respectively, which is significant. Our code has been made publicly available.

Hai Phan, Cindy Le, Vu Le et al.

Most face identification approaches employ a Siamese neural network to compare two images at the image embedding level. Yet, this technique can be subject to occlusion (e.g. faces with masks or sunglasses) and out-of-distribution data. DeepFace-EMD (Phan et al. 2022) reaches state-of-the-art accuracy on out-of-distribution data by first comparing two images at the image level, and then at the patch level. Yet, its later patch-wise re-ranking stage admits a large $O(n^3 \log n)$ time complexity (for $n$ patches in an image) due to the optimal transport optimization. In this paper, we propose a novel, 2-image Vision Transformers (ViTs) that compares two images at the patch level using cross-attention. After training on 2M pairs of images on CASIA Webface (Yi et al. 2014), our model performs at a comparable accuracy as DeepFace-EMD on out-of-distribution data, yet at an inference speed more than twice as fast as DeepFace-EMD (Phan et al. 2022). In addition, via a human study, our model shows promising explainability through the visualization of cross-attention. We believe our work can inspire more explorations in using ViTs for face identification.

Yihui He, Chenchen Zhu, Jianren Wang et al.

Large-scale object detection datasets (e.g., MS-COCO) try to define the ground truth bounding boxes as clear as possible. However, we observe that ambiguities are still introduced when labeling the bounding boxes. In this paper, we propose a novel bounding box regression loss for learning bounding box transformation and localization variance together. Our loss greatly improves the localization accuracies of various architectures with nearly no additional computation. The learned localization variance allows us to merge neighboring bounding boxes during non-maximum suppression (NMS), which further improves the localization performance. On MS-COCO, we boost the Average Precision (AP) of VGG-16 Faster R-CNN from 23.6% to 29.1%. More importantly, for ResNet-50-FPN Mask R-CNN, our method improves the AP and AP90 by 1.8% and 6.2% respectively, which significantly outperforms previous state-of-the-art bounding box refinement methods. Our code and models are available at: github.com/yihui-he/KL-Loss

Hanwen Zhang, Yao Liu, Peiyuan Jiang et al.

Explainable Multimodal Emotion Recognition plays a crucial role in applications such as human-computer interaction and social media analytics. However, current approaches struggle with cue-level perception and reasoning due to two main challenges: 1) general-purpose modality encoders are pretrained to capture global structures and general semantics rather than fine-grained emotional cues, resulting in limited sensitivity to emotional signals; and 2) available datasets usually involve a trade-off between annotation quality and scale, which leads to insufficient supervision for emotional cues and ultimately limits cue-level reasoning. Moreover, existing evaluation metrics are inadequate for assessing cue-level reasoning performance. To address these challenges, we propose eXplainable Emotion GPT (XEmoGPT), a novel EMER framework capable of both perceiving and reasoning over emotional cues. It incorporates two specialized modules: the Video Emotional Cue Bridge (VECB) and the Audio Emotional Cue Bridge (AECB), which enhance the video and audio encoders through carefully designed tasks for fine-grained emotional cue perception. To further support cue-level reasoning, we construct a large-scale dataset, EmoCue, designed to teach XEmoGPT how to reason over multimodal emotional cues. In addition, we introduce EmoCue-360, an automated metric that extracts and matches emotional cues using semantic similarity, and release EmoCue-Eval, a benchmark of 400 expert-annotated samples covering diverse emotional scenarios. Experimental results show that XEmoGPT achieves strong performance in both emotional cue perception and reasoning.

Chendi Lin, Heshan Liu, Qunshu Lin et al.

This paper presents Objaverse++, a curated subset of Objaverse enhanced with detailed attribute annotations by human experts. Recent advances in 3D content generation have been driven by large-scale datasets such as Objaverse, which contains over 800,000 3D objects collected from the Internet. Although Objaverse represents the largest available 3D asset collection, its utility is limited by the predominance of low-quality models. To address this limitation, we manually annotate 10,000 3D objects with detailed attributes, including aesthetic quality scores, texture color classifications, multi-object composition flags, transparency characteristics, etc. Then, we trained a neural network capable of annotating the tags for the rest of the Objaverse dataset. Through experiments and a user study on generation results, we demonstrate that models pre-trained on our quality-focused subset achieve better performance than those trained on the larger dataset of Objaverse in image-to-3D generation tasks. In addition, by comparing multiple subsets of training data filtered by our tags, our results show that the higher the data quality, the faster the training loss converges. These findings suggest that careful curation and rich annotation can compensate for the raw dataset size, potentially offering a more efficient path to develop 3D generative models. We release our enhanced dataset of approximately 500,000 curated 3D models to facilitate further research on various downstream tasks in 3D computer vision. In the near future, we aim to extend our annotations to cover the entire Objaverse dataset.

Thuan Than, Nhat-Anh Nguyen-Dang, Dung Nguyen et al.

Semi-Supervised Semantic Segmentation reduces reliance on extensive annotations by using unlabeled data and state-of-the-art models to improve overall performance. Despite the success of deep co-training methods, their underlying mechanisms remain underexplored. This work revisits Cross Pseudo Supervision with dual heterogeneous backbones and introduces Knowledge Consultation (SegKC) to further enhance segmentation performance. The proposed SegKC achieves significant improvements on Pascal and Cityscapes benchmarks, with mIoU scores of 87.1%, 89.2%, and 89.8% on Pascal VOC with the 1/4, 1/2, and full split partition, respectively, while maintaining a compact model architecture.

Cindy Le, Congrui Hetang, Chendi Lin et al.

We present EucliDreamer, a simple and effective method to generate textures for 3D models given text prompts and meshes. The texture is parametrized as an implicit function on the 3D surface, which is optimized with the Score Distillation Sampling (SDS) process and differentiable rendering. To generate high-quality textures, we leverage a depth-conditioned Stable Diffusion model guided by the depth image rendered from the mesh. We test our approach on 3D models in Objaverse and conducted a user study, which shows its superior quality compared to existing texturing methods like Text2Tex. In addition, our method converges 2 times faster than DreamFusion. Through text prompting, textures of diverse art styles can be produced. We hope Euclidreamer proides a viable solution to automate a labor-intensive stage in 3D content creation.

Jianren Wang, Yihui He

Visual object tracking (VOT) is an essential component for many applications, such as autonomous driving or assistive robotics. However, recent works tend to develop accurate systems based on more computationally expensive feature extractors for better instance matching. In contrast, this work addresses the importance of motion prediction in VOT. We use an off-the-shelf object detector to obtain instance bounding boxes. Then, a combination of camera motion decouple and Kalman filter is used for state estimation. Although our baseline system is a straightforward combination of standard methods, we obtain state-of-the-art results. Our method establishes new state-of-the-art performance on VOT (VOT-2016 and VOT-2018). Our proposed method improves the EAO on VOT-2016 from 0.472 of prior art to 0.505, from 0.410 to 0.431 on VOT-2018. To show the generalizability, we also test our method on video object segmentation (VOS: DAVIS-2016 and DAVIS-2017) and observe consistent improvement.

Yihui He, Rui Yan, Katerina Fragkiadaki et al.

A common approach to localize 3D human joints in a synchronized and calibrated multi-view setup consists of two-steps: (1) apply a 2D detector separately on each view to localize joints in 2D, and (2) perform robust triangulation on 2D detections from each view to acquire the 3D joint locations. However, in step 1, the 2D detector is limited to solving challenging cases which could potentially be better resolved in 3D, such as occlusions and oblique viewing angles, purely in 2D without leveraging any 3D information. Therefore, we propose the differentiable "epipolar transformer", which enables the 2D detector to leverage 3D-aware features to improve 2D pose estimation. The intuition is: given a 2D location p in the current view, we would like to first find its corresponding point p' in a neighboring view, and then combine the features at p' with the features at p, thus leading to a 3D-aware feature at p. Inspired by stereo matching, the epipolar transformer leverages epipolar constraints and feature matching to approximate the features at p'. Experiments on InterHand and Human3.6M show that our approach has consistent improvements over the baselines. Specifically, in the condition where no external data is used, our Human3.6M model trained with ResNet-50 backbone and image size 256 x 256 outperforms state-of-the-art by 4.23 mm and achieves MPJPE 26.9 mm.

Yihui He, Jianren Wang

Mistakes/uncertainties in object detection could lead to catastrophes when deploying robots in the real world. In this paper, we measure the uncertainties of object localization to minimize this kind of risk. Uncertainties emerge upon challenging cases like occlusion. The bounding box borders of an occluded object can have multiple plausible configurations. We propose a deep multivariate mixture of Gaussians model for probabilistic object detection. The covariances help to learn the relationship between the borders, and the mixture components potentially learn different configurations of an occluded part. Quantitatively, our model improves the AP of the baselines by 3.9% and 1.4% on CrowdHuman and MS-COCO respectively with almost no computational or memory overhead. Qualitatively, our model enjoys explainability since the resulting covariance matrices and the mixture components help measure uncertainties.

Yihui He, Jianing Qian, Jianren Wang et al.

Very deep convolutional neural networks (CNNs) have been firmly established as the primary methods for many computer vision tasks. However, most state-of-the-art CNNs are large, which results in high inference latency. Recently, depth-wise separable convolution has been proposed for image recognition tasks on computationally limited platforms such as robotics and self-driving cars. Though it is much faster than its counterpart, regular convolution, accuracy is sacrificed. In this paper, we propose a novel decomposition approach based on SVD, namely depth-wise decomposition, for expanding regular convolutions into depthwise separable convolutions while maintaining high accuracy. We show our approach can be further generalized to the multi-channel and multi-layer cases, based on Generalized Singular Value Decomposition (GSVD) [59]. We conduct thorough experiments with the latest ShuffleNet V2 model [47] on both random synthesized dataset and a large-scale image recognition dataset: ImageNet [10]. Our approach outperforms channel decomposition [73] on all datasets. More importantly, our approach improves the Top-1 accuracy of ShuffleNet V2 by ~2%.

Hai Phan, Dang Huynh, Yihui He et al.

MobileNet and Binary Neural Networks are two among the most widely used techniques to construct deep learning models for performing a variety of tasks on mobile and embedded platforms.In this paper, we present a simple yet efficient scheme to exploit MobileNet binarization at activation function and model weights. However, training a binary network from scratch with separable depth-wise and point-wise convolutions in case of MobileNet is not trivial and prone to divergence. To tackle this training issue, we propose a novel neural network architecture, namely MoBiNet - Mobile Binary Network in which skip connections are manipulated to prevent information loss and vanishing gradient, thus facilitate the training process. More importantly, while existing binary neural networks often make use of cumbersome backbones such as Alex-Net, ResNet, VGG-16 with float-type pre-trained weights initialization, our MoBiNet focuses on binarizing the already-compressed neural networks like MobileNet without the need of a pre-trained model to start with. Therefore, our proposal results in an effectively small model while keeping the accuracy comparable to existing ones. Experiments on ImageNet dataset show the potential of the MoBiNet as it achieves 54.40% top-1 accuracy and dramatically reduces the computational cost with binary operators.

Jianren Wang, Yihui He, Xiaobo Wang et al.

We introduce a prediction driven method for visual tracking and segmentation in videos. Instead of solely relying on matching with appearance cues for tracking, we build a predictive model which guides finding more accurate tracking regions efficiently. With the proposed prediction mechanism, we improve the model robustness against distractions and occlusions during tracking. We demonstrate significant improvements over state-of-the-art methods not only on visual tracking tasks (VOT 2016 and VOT 2018) but also on video segmentation datasets (DAVIS 2016 and DAVIS 2017).

Chenchen Zhu, Yihui He, Marios Savvides

We motivate and present feature selective anchor-free (FSAF) module, a simple and effective building block for single-shot object detectors. It can be plugged into single-shot detectors with feature pyramid structure. The FSAF module addresses two limitations brought up by the conventional anchor-based detection: 1) heuristic-guided feature selection; 2) overlap-based anchor sampling. The general concept of the FSAF module is online feature selection applied to the training of multi-level anchor-free branches. Specifically, an anchor-free branch is attached to each level of the feature pyramid, allowing box encoding and decoding in the anchor-free manner at an arbitrary level. During training, we dynamically assign each instance to the most suitable feature level. At the time of inference, the FSAF module can work jointly with anchor-based branches by outputting predictions in parallel. We instantiate this concept with simple implementations of anchor-free branches and online feature selection strategy. Experimental results on the COCO detection track show that our FSAF module performs better than anchor-based counterparts while being faster. When working jointly with anchor-based branches, the FSAF module robustly improves the baseline RetinaNet by a large margin under various settings, while introducing nearly free inference overhead. And the resulting best model can achieve a state-of-the-art 44.6% mAP, outperforming all existing single-shot detectors on COCO.

Devesh Walawalkar, Yihui He, Rohit Pillai

In this project, we worked on speech recognition, specifically predicting individual words based on both the video frames and audio. Empowered by convolutional neural networks, the recent speech recognition and lip reading models are comparable to human level performance. We re-implemented and made derivations of the state-of-the-art model. Then, we conducted rich experiments including the effectiveness of attention mechanism, more accurate residual network as the backbone with pre-trained weights and the sensitivity of our model with respect to audio input with/without noise.

Huasong Zhong, Xianggen Liu, Yihui He et al.

We propose a collection of three shift-based primitives for building efficient compact CNN-based networks. These three primitives (channel shift, address shift, shortcut shift) can reduce the inference time on GPU while maintains the prediction accuracy. These shift-based primitives only moves the pointer but avoids memory copy, thus very fast. For example, the channel shift operation is 12.7x faster compared to channel shuffle in ShuffleNet but achieves the same accuracy. The address shift and channel shift can be merged into the point-wise group convolution and invokes only a single kernel call, taking little time to perform spatial convolution and channel shift. Shortcut shift requires no time to realize residual connection through allocating space in advance. We blend these shift-based primitives with point-wise group convolution and built two inference-efficient CNN architectures named AddressNet and Enhanced AddressNet. Experiments on CIFAR100 and ImageNet datasets show that our models are faster and achieve comparable or better accuracy.

Yihui He, Ji Lin, Zhijian Liu et al.

Model compression is a critical technique to efficiently deploy neural network models on mobile devices which have limited computation resources and tight power budgets. Conventional model compression techniques rely on hand-crafted heuristics and rule-based policies that require domain experts to explore the large design space trading off among model size, speed, and accuracy, which is usually sub-optimal and time-consuming. In this paper, we propose AutoML for Model Compression (AMC) which leverage reinforcement learning to provide the model compression policy. This learning-based compression policy outperforms conventional rule-based compression policy by having higher compression ratio, better preserving the accuracy and freeing human labor. Under 4x FLOPs reduction, we achieved 2.7% better accuracy than the handcrafted model compression policy for VGG-16 on ImageNet. We applied this automated, push-the-button compression pipeline to MobileNet and achieved 1.81x speedup of measured inference latency on an Android phone and 1.43x speedup on the Titan XP GPU, with only 0.1% loss of ImageNet Top-1 accuracy.

Yihui He

We consider image classification with estimated depth. This problem falls into the domain of transfer learning, since we are using a model trained on a set of depth images to generate depth maps (additional features) for use in another classification problem using another disjoint set of images. It's challenging as no direct depth information is provided. Though depth estimation has been well studied, none have attempted to aid image classification with estimated depth. Therefore, we present a way of transferring domain knowledge on depth estimation to a separate image classification task over a disjoint set of train, and test data. We build a RGBD dataset based on RGB dataset and do image classification on it. Then evaluation the performance of neural networks on the RGBD dataset compared to the RGB dataset. From our experiments, the benefit is significant with shallow and deep networks. It improves ResNet-20 by 0.55% and ResNet-56 by 0.53%. Our code and dataset are available publicly.

Yihui He, Xiangyu Zhang, Jian Sun

In this paper, we introduce a new channel pruning method to accelerate very deep convolutional neural networks.Given a trained CNN model, we propose an iterative two-step algorithm to effectively prune each layer, by a LASSO regression based channel selection and least square reconstruction. We further generalize this algorithm to multi-layer and multi-branch cases. Our method reduces the accumulated error and enhance the compatibility with various architectures. Our pruned VGG-16 achieves the state-of-the-art results by 5x speed-up along with only 0.3% increase of error. More importantly, our method is able to accelerate modern networks like ResNet, Xception and suffers only 1.4%, 1.0% accuracy loss under 2x speed-up respectively, which is significant. Code has been made publicly available.

Yihui He, Ming Xiang

With applications to many disciplines, the traveling salesman problem (TSP) is a classical computer science optimization problem with applications to industrial engineering, theoretical computer science, bioinformatics, and several other disciplines. In recent years, there have been a plethora of novel approaches for approximate solutions ranging from simplistic greedy to cooperative distributed algorithms derived from artificial intelligence. In this paper, we perform an evaluation and analysis of cornerstone algorithms for the Euclidean TSP. We evaluate greedy, 2-opt, and genetic algorithms. We use several datasets as input for the algorithms including a small dataset, a mediumsized dataset representing cities in the United States, and a synthetic dataset consisting of 200 cities to test algorithm scalability. We discover that the greedy and 2-opt algorithms efficiently calculate solutions for smaller datasets. Genetic algorithm has the best performance for optimality for medium to large datasets, but generally have longer runtime. Our implementations is public available.

Yihui He, Xiaobo Ma, Xiapu Luo et al.

Security surveillance is one of the most important issues in smart cities, especially in an era of terrorism. Deploying a number of (video) cameras is a common surveillance approach. Given the never-ending power offered by vehicles to metropolises, exploiting vehicle traffic to design camera placement strategies could potentially facilitate security surveillance. This article constitutes the first effort toward building the linkage between vehicle traffic and security surveillance, which is a critical problem for smart cities. We expect our study could influence the decision making of surveillance camera placement, and foster more research of principled ways of security surveillance beneficial to our physical-world life. Code has been made publicly available.

Yudong Liang, Ze Yang, Kai Zhang et al.

Recent years have witnessed great success of convolutional neural network (CNN) for various problems both in low and high level visions. Especially noteworthy is the residual network which was originally proposed to handle high-level vision problems and enjoys several merits. This paper aims to extend the merits of residual network, such as skip connection induced fast training, for a typical low-level vision problem, i.e., single image super-resolution. In general, the two main challenges of existing deep CNN for supper-resolution lie in the gradient exploding/vanishing problem and large numbers of parameters or computational cost as CNN goes deeper. Correspondingly, the skip connections or identity mapping shortcuts are utilized to avoid gradient exploding/vanishing problem. In addition, the skip connections have naturally centered the activation which led to better performance. To tackle with the second problem, a lightweight CNN architecture which has carefully designed width, depth and skip connections was proposed. In particular, a strategy of gradually varying the shape of network has been proposed for residual network. Different residual architectures for image super-resolution have also been compared. Experimental results have demonstrated that the proposed CNN model can not only achieve state-of-the-art PSNR and SSIM results for single image super-resolution but also produce visually pleasant results. This paper has extended the mmm 2017 oral conference paper with a considerable new analyses and more experiments especially from the perspective of centering activations and ensemble behaviors of residual network.