159 citations · 943 across the 39 of their papers we have counts for
39 papers
PhaseLoRA: Control-Regime-Conditioned Low-Rank Adaptation for Continuous-Action Vision-Language-Action Policies
Yufei Guo, Yinan Wu, Haoran Duan +2
Parameter-efficient fine-tuning (PEFT) is a natural way to adapt pretrained vision-language-action (VLA) policies, but most adapter designs apply temporally static updates througho…
Confidence-based Visual Dispersal for Few-shot Unsupervised Domain Adaptation
Yizhe Xiong, Hui Chen, Zijia Lin +2
Unsupervised domain adaptation aims to transfer knowledge from a fully-labeled source domain to an unlabeled target domain. However, in real-world scenarios, providing abundant lab…
CAIT: Triple-Win Compression towards High Accuracy, Fast Inference, and Favorable Transferability For ViTs
Ao Wang, Hui Chen, Zijia Lin +3
Vision Transformers (ViTs) have emerged as state-of-the-art models for various vision tasks recently. However, their heavy computation costs remain daunting for resource-limited de…
RepViT: Revisiting Mobile CNN From ViT Perspective
Ao Wang, Hui Chen, Zijia Lin +2
Recently, lightweight Vision Transformers (ViTs) demonstrate superior performance and lower latency, compared with lightweight Convolutional Neural Networks (CNNs), on resource-con…
Consolidator: Mergeable Adapter with Grouped Connections for Visual Adaptation
Tianxiang Hao, Hui Chen, Yuchen Guo +1
Recently, transformers have shown strong ability as visual feature extractors, surpassing traditional convolution-based models in various scenarios. However, the success of vision…
Box-Level Active Detection
Mengyao Lyu, Jundong Zhou, Hui Chen +7
Active learning selects informative samples for annotation within budget, which has proven efficient recently on object detection. However, the widely used active detection benchma…
X-ReID: Cross-Instance Transformer for Identity-Level Person Re-Identification
Leqi Shen, Tao He, Yuchen Guo +1
Currently, most existing person re-identification methods use Instance-Level features, which are extracted only from a single image. However, these Instance-Level features can easi…
Confidence-guided Centroids for Unsupervised Person Re-Identification
Yunqi Miao, Jiankang Deng, Guiguang Ding +1
Unsupervised person re-identification (ReID) aims to train a feature extractor for identity retrieval without exploiting identity labels. Due to the blind trust in imperfect cluste…
Ground Plane Matters: Picking Up Ground Plane Prior in Monocular 3D Object Detection
Fan Yang, Xinhao Xu, Hui Chen +4
The ground plane prior is a very informative geometry clue in monocular 3D object detection (M3OD). However, it has been neglected by most mainstream methods. In this paper, we ide…
JDRec: Practical Actor-Critic Framework for Online Combinatorial Recommender System
Xin Zhao, Zhiwei Fang, Yuchen Guo +3
A combinatorial recommender (CR) system feeds a list of items to a user at a time in the result page, in which the user behavior is affected by both contextual information and item…
A High-Accuracy Unsupervised Person Re-identification Method Using Auxiliary Information Mined from Datasets
Hehan Teng, Tao He, Yuchen Guo +1
Supervised person re-identification methods rely heavily on high-quality cross-camera training label. This significantly hinders the deployment of re-ID models in real-world applic…
Re-parameterizing Your Optimizers rather than Architectures
Xiaohan Ding, Honghao Chen, Xiangyu Zhang +3
The well-designed structures in neural networks reflect the prior knowledge incorporated into the models. However, though different models have various priors, we are used to train…
A Free Lunch to Person Re-identification: Learning from Automatically Generated Noisy Tracklets
Hehan Teng, Tao He, Yuchen Guo +2
A series of unsupervised video-based re-identification (re-ID) methods have been proposed to solve the problem of high labor cost required to annotate re-ID datasets. But their per…
Scaling Up Your Kernels to 31x31: Revisiting Large Kernel Design in CNNs
Xiaohan Ding, Xiangyu Zhang, Yizhuang Zhou +3
We revisit large kernel design in modern convolutional neural networks (CNNs). Inspired by recent advances in vision transformers (ViTs), in this paper, we demonstrate that using a…
TAGPerson: A Target-Aware Generation Pipeline for Person Re-identification
Kai Chen, Weihua Chen, Tao He +5
Nowadays, real data in person re-identification (ReID) task is facing privacy issues, e.g., the banned dataset DukeMTMC-ReID. Thus it becomes much harder to collect real data for R…
RepMLPNet: Hierarchical Vision MLP with Re-parameterized Locality
Xiaohan Ding, Honghao Chen, Xiangyu Zhang +2
Compared to convolutional layers, fully-connected (FC) layers are better at modeling the long-range dependencies but worse at capturing the local patterns, hence usually less favor…
LODE: Deep Local Deblurring and A New Benchmark
Zerun Wang, Liuyu Xiang, Fan Yang +6
While recent deep deblurring algorithms have achieved remarkable progress, most existing methods focus on the global deblurring problem, where the image blur mostly arises from sev…
Emotion Recognition from Multiple Modalities: Fundamentals and Methodologies
Sicheng Zhao, Guoli Jia, Jufeng Yang +2
Humans are emotional creatures. Multiple modalities are often involved when we express emotions, whether we do so explicitly (e.g., facial expression, speech) or implicitly (e.g.,…
Manipulating Identical Filter Redundancy for Efficient Pruning on Deep and Complicated CNN
Xiaohan Ding, Tianxiang Hao, Jungong Han +2
The existence of redundancy in Convolutional Neural Networks (CNNs) enables us to remove some filters/channels with acceptable performance drops. However, the training objective of…
Affective Image Content Analysis: Two Decades Review and New Perspectives
Sicheng Zhao, Xingxu Yao, Jufeng Yang +5
Images can convey rich semantics and induce various emotions in viewers. Recently, with the rapid advancement of emotional intelligence and the explosive growth of visual data, ext…
RepMLP: Re-parameterizing Convolutions into Fully-connected Layers for Image Recognition
Xiaohan Ding, Chunlong Xia, Xiangyu Zhang +3
We propose RepMLP, a multi-layer-perceptron-style neural network building block for image recognition, which is composed of a series of fully-connected (FC) layers. Compared to con…
Diverse Branch Block: Building a Convolution as an Inception-like Unit
Xiaohan Ding, Xiangyu Zhang, Jungong Han +1
We propose a universal building block of Convolutional Neural Network (ConvNet) to improve the performance without any inference-time costs. The block is named Diverse Branch Block…
Computational Emotion Analysis From Images: Recent Advances and Future Directions
Sicheng Zhao, Quanwei Huang, Youbao Tang +4
Emotions are usually evoked in humans by images. Recently, extensive research efforts have been dedicated to understanding the emotions of images. In this chapter, we aim to introd…
RepVGG: Making VGG-style ConvNets Great Again
Xiaohan Ding, Xiangyu Zhang, Ningning Ma +3
We present a simple but powerful architecture of convolutional neural network, which has a VGG-like inference-time body composed of nothing but a stack of 3x3 convolution and ReLU,…