1 code implementation • 23 May 2024 • Ao Wang, Hui Chen, Lihao Liu, Kai Chen, Zijia Lin, Jungong Han, Guiguang Ding
In this work, we aim to further advance the performance-efficiency boundary of YOLOs from both the post-processing and model architecture.
no code implementations • 1 May 2024 • Sicheng Zhao, Hui Chen, Hu Huang, Pengfei Xu, Guiguang Ding
Domain adaptation (DA) aims to address this problem by aligning the distributions between the source and target domains.
no code implementations • 27 Apr 2024 • Yizhe Xiong, Xiansheng Chen, Xin Ye, Hui Chen, Zijia Lin, Haoran Lian, Jianwei Niu, Guiguang Ding
We first investigate the imbalance of loss on each token positions and develop a reciprocal-law across model scales and training stages.
no code implementations • 27 Apr 2024 • Haoran Lian, Yizhe Xiong, Jianwei Niu, Shasha Mo, Zhenpeng Su, Zijia Lin, Peng Liu, Hui Chen, Guiguang Ding
Due to their infrequent appearance in the text corpus, Scaffold Tokens pose a learning imbalance issue for language models.
no code implementations • 14 Mar 2024 • Yizhe Xiong, Hui Chen, Tianxiang Hao, Zijia Lin, Jungong Han, Yuesong Zhang, Guoxin Wang, Yongjun Bao, Guiguang Ding
Consequently, a simple combination of them cannot guarantee accomplishing both training efficiency and inference efficiency with minimal costs.
no code implementations • 29 Feb 2024 • Juexiao Feng, Yuhong Yang, Yanchun Xie, Yaqian Li, Yandong Guo, Yuchen Guo, Yuwei He, Liuyu Xiang, Guiguang Ding
In recent years, object detection in deep learning has experienced rapid development.
1 code implementation • 26 Dec 2023 • Mengyao Lyu, Yuhong Yang, Haiwen Hong, Hui Chen, Xuan Jin, Yuan He, Hui Xue, Jungong Han, Guiguang Ding
The prevalent use of commercial and open-source diffusion models (DMs) for text-to-image generation prompts risk mitigation to prevent undesired behaviors.
no code implementations • 17 Dec 2023 • Tianxiang Hao, Mengyao Lyu, Hui Chen, Sicheng Zhao, Jungong Han, Guiguang Ding
On the other hand, complicated structures and update rules largely increase the computation and storage cost.
2 code implementations • 10 Dec 2023 • Ao Wang, Hui Chen, Zijia Lin, Jungong Han, Guiguang Ding
Here, to achieve real-time segmenting anything on mobile devices, following MobileSAM, we replace the heavyweight image encoder in SAM with RepViT model, ending up with the RepViT-SAM model.
1 code implementation • 30 Oct 2023 • Zhenpeng Su, Xing Wu, Xue Bai, Zijia Lin, Hui Chen, Guiguang Ding, Wei Zhou, Songlin Hu
Experiments reveal that models incorporating the proposed MiLe Loss can gain consistent performance improvement on downstream benchmarks.
Ranked #90 on Multi-task Language Understanding on MMLU
no code implementations • 27 Sep 2023 • Ao Wang, Hui Chen, Zijia Lin, Sicheng Zhao, Jungong Han, Guiguang Ding
We further employ a consistent dynamic channel pruning (CDCP) strategy to dynamically prune unimportant channels in ViTs.
1 code implementation • ICCV 2023 • Yizhe Xiong, Hui Chen, Zijia Lin, Sicheng Zhao, Guiguang Ding
To address this issue, recent works consider the Few-shot Unsupervised Domain Adaptation (FUDA) where only a few source samples are labeled, and conduct knowledge transfer via self-supervised learning methods.
7 code implementations • 18 Jul 2023 • Ao Wang, Hui Chen, Zijia Lin, Jungong Han, Guiguang Ding
Recently, lightweight Vision Transformers (ViTs) demonstrate superior performance and lower latency, compared with lightweight Convolutional Neural Networks (CNNs), on resource-constrained mobile devices.
1 code implementation • 30 Apr 2023 • Tianxiang Hao, Hui Chen, Yuchen Guo, Guiguang Ding
To further enhance the model's capacity to transfer knowledge under a constrained storage budget and keep inference efficient, we consolidate the parameters in two stages: 1. between adaptation and storage, and 2. between loading and inference.
1 code implementation • CVPR 2023 • Mengyao Lyu, Jundong Zhou, Hui Chen, YiJie Huang, Dongdong Yu, Yaqian Li, Yandong Guo, Yuchen Guo, Liuyu Xiang, Guiguang Ding
Active learning selects informative samples for annotation within budget, which has proven efficient recently on object detection.
no code implementations • 4 Feb 2023 • Leqi Shen, Tao He, Yuchen Guo, Guiguang Ding
In this paper, we propose to promote Instance-Level features to Identity-Level features by employing cross-attention to incorporate information from one image to another of the same identity, thus more unified and discriminative pedestrian information can be obtained.
no code implementations • 22 Nov 2022 • Yunqi Miao, Jiankang Deng, Guiguang Ding, Jungong Han
Since samples with high confidence are exclusively involved in the formation of centroids, the identity information of low-confidence samples, i. e., boundary samples, are NOT likely to contribute to the corresponding centroid.
no code implementations • 3 Nov 2022 • Fan Yang, Xinhao Xu, Hui Chen, Yuchen Guo, Jungong Han, Kai Ni, Guiguang Ding
To pick up the ground plane prior for M3OD, we propose a Ground Plane Enhanced Network (GPENet) which resolves both issues at one go.
1 code implementation • 30 May 2022 • Xiaohan Ding, Honghao Chen, Xiangyu Zhang, Kaiqi Huang, Jungong Han, Guiguang Ding
For the extreme simplicity of model structure, we focus on a VGG-style plain model and showcase that such a simple model trained with a RepOptimizer, which is referred to as RepOpt-VGG, performs on par with or better than the recent well-designed models.
1 code implementation • 6 May 2022 • Hehan Teng, Tao He, Yuchen Guo, Guiguang Ding
Combined with auxiliary information exploiting modules, our methods achieve mAP of 89. 9% on DukeMTMC, where TOC, STS and SCP all contributed considerable performance improvements.
Ranked #1 on Unsupervised Person Re-Identification on MARS
no code implementations • 2 Apr 2022 • Hehan Teng, Tao He, Yuchen Guo, Zhenhua Guo, Guiguang Ding
Extensive experiments on MARS with various manually generated noises show the effectiveness of the proposed framework.
7 code implementations • CVPR 2022 • Xiaohan Ding, Xiangyu Zhang, Yizhuang Zhou, Jungong Han, Guiguang Ding, Jian Sun
We revisit large kernel design in modern convolutional neural networks (CNNs).
Ranked #75 on Image Classification on <h2>oi</h2>
1 code implementation • 28 Dec 2021 • Kai Chen, Weihua Chen, Tao He, Rong Du, Fan Wang, Xiuyu Sun, Yuchen Guo, Guiguang Ding
In TAGPerson, we extract information from target scenes and use them to control our parameterized rendering process to generate target-aware synthetic images, which would hold a smaller gap to the real images in the target domain.
4 code implementations • CVPR 2022 • Xiaohan Ding, Honghao Chen, Xiangyu Zhang, Jungong Han, Guiguang Ding
Our results reveal that 1) Locality Injection is a general methodology for MLP models; 2) RepMLPNet has favorable accuracy-efficiency trade-off compared to the other MLPs; 3) RepMLPNet is the first MLP that seamlessly transfer to Cityscapes semantic segmentation.
Ranked #61 on Semantic Segmentation on Cityscapes val
no code implementations • 29 Sep 2021 • Tao He, Tongkun Xu, Weihua Chen, Yuchen Guo, Guiguang Ding, Zhenhua Guo
Due to the discrepancies between cameras caused by illumination, background, or viewpoint, the underlying difficulty for Re-ID is the camera bias problem, which leads to the large gap of within-identity features from different cameras.
1 code implementation • 19 Sep 2021 • Zerun Wang, Liuyu Xiang, Fan Yang, Jinzhao Qian, Jie Hu, Haidong Huang, Jungong Han, Yuchen Guo, Guiguang Ding
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 severe camera shake.
no code implementations • 18 Aug 2021 • Sicheng Zhao, Guoli Jia, Jufeng Yang, Guiguang Ding, Kurt Keutzer
In this tutorial, we discuss several key aspects of multi-modal emotion recognition (MER).
2 code implementations • 30 Jul 2021 • Xiaohan Ding, Tianxiang Hao, Jungong Han, Yuchen Guo, Guiguang Ding
The existence of redundancy in Convolutional Neural Networks (CNNs) enables us to remove some filters/channels with acceptable performance drops.
no code implementations • 30 Jun 2021 • Sicheng Zhao, Xingxu Yao, Jufeng Yang, Guoli Jia, Guiguang Ding, Tat-Seng Chua, Björn W. Schuller, Kurt Keutzer
Images can convey rich semantics and induce various emotions in viewers.
10 code implementations • 5 May 2021 • Xiaohan Ding, Chunlong Xia, Xiangyu Zhang, Xiaojie Chu, Jungong Han, Guiguang Ding
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.
Ranked #754 on Image Classification on <h2>oi</h2>
2 code implementations • CVPR 2021 • Xiaohan Ding, Xiangyu Zhang, Jungong Han, Guiguang Ding
We propose a universal building block of Convolutional Neural Network (ConvNet) to improve the performance without any inference-time costs.
no code implementations • 19 Mar 2021 • Sicheng Zhao, Quanwei Huang, YouBao Tang, Xingxu Yao, Jufeng Yang, Guiguang Ding, Björn W. Schuller
Recently, extensive research efforts have been dedicated to understanding the emotions of images.
2 code implementations • 14 Jan 2021 • Xin He, Shihao Wang, Xiaowen Chu, Shaohuai Shi, Jiangping Tang, Xin Liu, Chenggang Yan, Jiyong Zhang, Guiguang Ding
The experimental results show that our automatically searched models (CovidNet3D) outperform the baseline human-designed models on the three datasets with tens of times smaller model size and higher accuracy.
23 code implementations • CVPR 2021 • Xiaohan Ding, Xiangyu Zhang, Ningning Ma, Jungong Han, Guiguang Ding, Jian Sun
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, while the training-time model has a multi-branch topology.
Ranked #44 on Semantic Segmentation on Cityscapes val
no code implementations • 25 Nov 2020 • Sicheng Zhao, Xuanbai Chen, Xiangyu Yue, Chuang Lin, Pengfei Xu, Ravi Krishna, Jufeng Yang, Guiguang Ding, Alberto L. Sangiovanni-Vincentelli, Kurt Keutzer
First, we generate an adapted domain to align the source and target domains on the pixel-level by improving CycleGAN with a multi-scale structured cycle-consistency loss.
6 code implementations • ICCV 2021 • Xiaohan Ding, Tianxiang Hao, Jianchao Tan, Ji Liu, Jungong Han, Yuchen Guo, Guiguang Ding
Via training with regular SGD on the former but a novel update rule with penalty gradients on the latter, we realize structured sparsity.
no code implementations • 17 Jun 2020 • Yunqi Miao, Zijia Lin, Guiguang Ding, Jungong Han
In this paper, we propose a Shallow feature based Dense Attention Network (SDANet) for crowd counting from still images, which diminishes the impact of backgrounds via involving a shallow feature based attention model, and meanwhile, captures multi-scale information via densely connecting hierarchical image features.
1 code implementation • IEEE Transactions on Cybernetics 2020 • Hancheng Zhu, Leida Li, Jinjian Wu, Sicheng Zhao, Guiguang Ding, and Guangming Shi
Typical image aesthetics assessment (IAA) is modeled for the generic aesthetics perceived by an ``average'' user.
no code implementations • CVPR 2020 • Xueyang Wang, Xiya Zhang, Yinheng Zhu, Yuchen Guo, Xiaoyun Yuan, Liuyu Xiang, Zerun Wang, Guiguang Ding, David J. Brady, Qionghai Dai, Lu Fang
We believe PANDA will contribute to the community of artificial intelligence and praxeology by understanding human behaviors and interactions in large-scale real-world scenes.
1 code implementation • CVPR 2020 • Hui Chen, Guiguang Ding, Xudong Liu, Zijia Lin, Ji Liu, Jungong Han
Existing methods leverage the attention mechanism to explore such correspondence in a fine-grained manner.
Ranked #19 on Cross-Modal Retrieval on Flickr30k
1 code implementation • ECCV 2020 • Liuyu Xiang, Guiguang Ding, Jungong Han
We refer to these models as 'Experts', and the proposed LFME framework aggregates the knowledge from multiple 'Experts' to learn a unified student model.
Ranked #26 on Long-tail Learning on Places-LT
4 code implementations • NeurIPS 2019 • Xiaohan Ding, Guiguang Ding, Xiangxin Zhou, Yuchen Guo, Jungong Han, Ji Liu
Deep Neural Network (DNN) is powerful but computationally expensive and memory intensive, thus impeding its practical usage on resource-constrained front-end devices.
1 code implementation • 11 Sep 2019 • Sicheng Zhao, Zizhou Jia, Hui Chen, Leida Li, Guiguang Ding, Kurt Keutzer
By optimizing the PCR loss, PDANet can generate a polarity preserved attention map and thus improve the emotion regression performance.
5 code implementations • ICCV 2019 • Xiaohan Ding, Yuchen Guo, Guiguang Ding, Jungong Han
We propose Asymmetric Convolution Block (ACB), an architecture-neutral structure as a CNN building block, which uses 1D asymmetric convolutions to strengthen the square convolution kernels.
1 code implementation • 12 Jul 2019 • Hui Chen, Zijia Lin, Guiguang Ding, JianGuang Lou, Yusen Zhang, Borje Karlsson
The dominant approaches for named entity recognition (NER) mostly adopt complex recurrent neural networks (RNN), e. g., long-short-term-memory (LSTM).
Ranked #23 on Named Entity Recognition (NER) on Ontonotes v5 (English)
no code implementations • 2 Jun 2019 • Liuyu Xiang, Xiaoming Jin, Guiguang Ding, Jungong Han, Leida Li
Pedestrian attribute recognition has received increasing attention due to its important role in video surveillance applications.
no code implementations • 28 May 2019 • Liuyu Xiang, Xiaoming Jin, Lan Yi, Guiguang Ding
Deep learning models such as convolutional neural networks and recurrent networks are widely applied in text classification.
1 code implementation • 12 May 2019 • Xiaohan Ding, Guiguang Ding, Yuchen Guo, Jungong Han, Chenggang Yan
It is not easy to design and run Convolutional Neural Networks (CNNs) due to: 1) finding the optimal number of filters (i. e., the width) at each layer is tricky, given an architecture; and 2) the computational intensity of CNNs impedes the deployment on computationally limited devices.
1 code implementation • CVPR 2019 • Xiaohan Ding, Guiguang Ding, Yuchen Guo, Jungong Han
The redundancy is widely recognized in Convolutional Neural Networks (CNNs), which enables to remove unimportant filters from convolutional layers so as to slim the network with acceptable performance drop.
no code implementations • CVPR 2017 • Yang Long, Li Liu, Ling Shao, Fumin Shen, Guiguang Ding, Jungong Han
Using the proposed Unseen Visual Data Synthesis (UVDS) algorithm, semantic attributes are effectively utilised as an intermediate clue to synthesise unseen visual features at the training stage.
no code implementations • CVPR 2015 • Zijia Lin, Guiguang Ding, Mingqing Hu, Jian-Min Wang
With benefits of low storage costs and high query speeds, hashing methods are widely researched for efficiently retrieving large-scale data, which commonly contains multiple views, e. g. a news report with images, videos and texts.
no code implementations • CVPR 2014 • Mingsheng Long, Jian-Min Wang, Guiguang Ding, Jiaguang Sun, Philip S. Yu
Visual domain adaptation, which learns an accurate classifier for a new domain using labeled images from an old domain, has shown promising value in computer vision yet still been a challenging problem.
no code implementations • CVPR 2014 • Guiguang Ding, Yuchen Guo, Jile Zhou
In this paper, we study the problems of learning hash functions in the context of multimodal data for cross-view similarity search.
no code implementations • CVPR 2013 • Zijia Lin, Guiguang Ding, Mingqing Hu, Jian-Min Wang, Xiaojun Ye
Though widely utilized for facilitating image management, user-provided image tags are usually incomplete and insufficient to describe the whole semantic content of corresponding images, resulting in performance degradations in tag-dependent applications and thus necessitating effective tag completion methods.
no code implementations • CVPR 2013 • Mingsheng Long, Guiguang Ding, Jian-Min Wang, Jiaguang Sun, Yuchen Guo, Philip S. Yu
In this paper, we propose a Transfer Sparse Coding (TSC) approach to construct robust sparse representations for classifying cross-distribution images accurately.