1 code implementation • 21 Jun 2023 • Xu Zhao, Wenchao Ding, Yongqi An, Yinglong Du, Tao Yu, Min Li, Ming Tang, Jinqiao Wang
In this paper, we propose a speed-up alternative method for this fundamental task with comparable performance.
Ranked #4 on Zero-Shot Instance Segmentation on LVIS v1.0 val
1 code implementation • 29 Aug 2023 • Zhaopeng Gu, Bingke Zhu, Guibo Zhu, Yingying Chen, Ming Tang, Jinqiao Wang
Large Vision-Language Models (LVLMs) such as MiniGPT-4 and LLaVA have demonstrated the capability of understanding images and achieved remarkable performance in various visual tasks.
2 code implementations • 1 Jul 2021 • Jing Liu, Xinxin Zhu, Fei Liu, Longteng Guo, Zijia Zhao, Mingzhen Sun, Weining Wang, Hanqing Lu, Shiyu Zhou, Jiajun Zhang, Jinqiao Wang
In this paper, we propose an Omni-perception Pre-Trainer (OPT) for cross-modal understanding and generation, by jointly modeling visual, text and audio resources.
Ranked #1 on Image Retrieval on Localized Narratives
1 code implementation • 30 Jul 2021 • Zhiyang Chen, Yousong Zhu, Chaoyang Zhao, Guosheng Hu, Wei Zeng, Jinqiao Wang, Ming Tang
To address this problem, we propose a new Deformable Patch (DePatch) module which learns to adaptively split the images into patches with different positions and scales in a data-driven way rather than using predefined fixed patches.
Ranked #17 on Semantic Segmentation on DensePASS
1 code implementation • 2 Nov 2023 • Jianghao Chen, Pu Jian, Tengxiao Xi, Dongyi Yi, Qianlong Du, Chenglin Ding, Guibo Zhu, Chengqing Zong, Jinqiao Wang, Jiajun Zhang
Using our proposed approach, we release the largest and latest large-scale high-quality Chinese web text ChineseWebText, which consists of 1. 42 TB and each text is associated with a quality score, facilitating the LLM researchers to choose the data according to the desired quality thresholds.
1 code implementation • ECCV 2020 • Kuan Zhu, Haiyun Guo, Zhiwei Liu, Ming Tang, Jinqiao Wang
In this paper, we propose the identity-guided human semantic parsing approach (ISP) to locate both the human body parts and personal belongings at pixel-level for aligned person re-ID only with person identity labels.
Ranked #40 on Person Re-Identification on DukeMTMC-reID
2 code implementations • 28 Sep 2022 • Zhiyang Chen, Yousong Zhu, Zhaowen Li, Fan Yang, Wei Li, Haixin Wang, Chaoyang Zhao, Liwei Wu, Rui Zhao, Jinqiao Wang, Ming Tang
Obj2Seq is able to flexibly determine input categories to satisfy customized requirements, and be easily extended to different visual tasks.
1 code implementation • CVPR 2021 • Tong Wang, Yousong Zhu, Chaoyang Zhao, Wei Zeng, Jinqiao Wang, Ming Tang
To address the problem of long-tail distribution for the large vocabulary object detection task, existing methods usually divide the whole categories into several groups and treat each group with different strategies.
1 code implementation • 24 Nov 2023 • Yufei Zhan, Yousong Zhu, Zhiyang Chen, Fan Yang, Ming Tang, Jinqiao Wang
More importantly, we present $\textbf{Griffon}$, a purely LVLM-based baseline, which does not require the introduction of any special tokens, expert models, or additional detection modules.
1 code implementation • 14 Mar 2024 • Yufei Zhan, Yousong Zhu, Hongyin Zhao, Fan Yang, Ming Tang, Jinqiao Wang
Large Vision Language Models have achieved fine-grained object perception, but the limitation of image resolution remains a significant obstacle to surpass the performance of task-specific experts in complex and dense scenarios.
3 code implementations • ICCV 2017 • Yousong Zhu, Chaoyang Zhao, Jinqiao Wang, Xu Zhao, Yi Wu, Hanqing Lu
To fully explore the local and global properties, in this paper, we propose a novel fully convolutional network, named as CoupleNet, to couple the global structure with local parts for object detection.
Ranked #5 on Object Detection on PASCAL VOC 2007
1 code implementation • 8 Mar 2022 • Kuan Zhu, Haiyun Guo, Tianyi Yan, Yousong Zhu, Jinqiao Wang, Ming Tang
PASS learns to match the output of the local views and global views on the same [PART].
1 code implementation • CVPR 2023 • Yongqi An, Xu Zhao, Tao Yu, Haiyun Guo, Chaoyang Zhao, Ming Tang, Jinqiao Wang
However, previous unsupervised deep learning BGS algorithms perform poorly in sophisticated scenarios such as shadows or night lights, and they cannot detect objects outside the pre-defined categories.
1 code implementation • 26 Jul 2017 • Bingke Zhu, Yingying Chen, Jinqiao Wang, Si Liu, Bo Zhang, Ming Tang
Finally, an automatic portrait animation system based on fast deep matting is built on mobile devices, which does not need any interaction and can realize real-time matting with 15 fps.
1 code implementation • 2 Dec 2022 • Fangxun Shu, Biaolong Chen, Yue Liao, Shuwen Xiao, Wenyu Sun, Xiaobo Li, Yousong Zhu, Jinqiao Wang, Si Liu
Our MAC aims to reduce video representation's spatial and temporal redundancy in the VidLP model by a mask sampling mechanism to improve pre-training efficiency.
Ranked #36 on Video Retrieval on MSR-VTT-1kA (using extra training data)
1 code implementation • 19 Dec 2023 • Yongqi An, Xu Zhao, Tao Yu, Ming Tang, Jinqiao Wang
Retraining-free is important for LLMs' pruning methods.
1 code implementation • 14 Oct 2020 • Xiaoqing Liang, Xu Zhao, Chaoyang Zhao, Nanfei Jiang, Ming Tang, Jinqiao Wang
This method decouples the distillation task of face detection into two subtasks, i. e., the classification distillation subtask and the regression distillation subtask.
1 code implementation • 18 Jan 2022 • Nanfei Jiang, Xu Zhao, Chaoyang Zhao, Yongqi An, Ming Tang, Jinqiao Wang
MaskSparsity imposes the fine-grained sparse regularization on the specific filters selected by a pruning mask, rather than all the filters of the model.
1 code implementation • 25 Feb 2023 • Jinzhao Luo, Lu Zhou, Guibo Zhu, Guojing Ge, Beiying Yang, Jinqiao Wang
Most current methods adopt graph convolutional network (GCN) for topology modeling, but GCN-based methods are limited in long-distance correlation modeling and generalizability.
no code implementations • CVPR 2018 • Ming Tang, Bin Yu, Fan Zhang, Jinqiao Wang
In this paper, we will introduce the MKL into KCF in a different way than MKCF.
no code implementations • 17 Jun 2018 • Ming Tang, Linyu Zheng, Bin Yu, Jinqiao Wang
To achieve the fast training and detection, a set of cyclic bases is introduced to construct the filter.
no code implementations • 25 Nov 2017 • Jinqiao Wang, Ming Tang, Linyu Zheng, Jiayi Feng
In recent years, two types of trackers, namely correlation filter based tracker (CF tracker) and structured output tracker (Struck), have exhibited the state-of-the-art performance.
no code implementations • 13 Sep 2017 • Yunze Gao, Yingying Chen, Jinqiao Wang, Hanqing Lu
Reading text in the wild is a challenging task in the field of computer vision.
no code implementations • 24 Jul 2017 • Xu Zhao, Yingying Chen, Ming Tang, Jinqiao Wang
In the first stage, a convolutional encoder-decoder sub-network is employed to reconstruct the background images and encode rich prior knowledge of background scenes.
no code implementations • 18 Dec 2018 • Yunze Gao, Yingying Chen, Jinqiao Wang, Zhen Lei, Xiao-Yu Zhang, Hanqing Lu
In this paper, we propose a novel Recurrent Calibration Network (RCN) for irregular scene text recognition.
no code implementations • CVPR 2017 • Zhen Wei, Yao Sun, Jinqiao Wang, Hanjiang Lai, Si Liu
In this paper, we introduce a novel approach to regulate receptive field in deep image parsing network automatically.
no code implementations • ICCV 2015 • Jianlong Fu, Yue Wu, Tao Mei, Jinqiao Wang, Hanqing Lu, Yong Rui
The development of deep learning has empowered machines with comparable capability of recognizing limited image categories to human beings.
no code implementations • CVPR 2019 • Zhiwei Liu, Xiangyu Zhu, Guosheng Hu, Haiyun Guo, Ming Tang, Zhen Lei, Neil M. Robertson, Jinqiao Wang
Despite this, we notice that the semantic ambiguity greatly degrades the detection performance.
Ranked #1 on Face Alignment on 300W (NME_inter-pupil (%, Full) metric)
no code implementations • ECCV 2020 • Linyu Zheng, Ming Tang, Yingying Chen, Jinqiao Wang, Hanqing Lu
After observing that the features used in most online discriminatively trained trackers are not optimal, in this paper, we propose a novel and effective architecture to learn optimal feature embeddings for online discriminative tracking.
no code implementations • ECCV 2020 • Lu Zhou, Yingying Chen, Yunze Gao, Jinqiao Wang, Hanqing Lu
To overcome the defects caused by the erasing operation, we perform feature reconstruction to recover the information destroyed by occlusion and details lost in cleaning procedure.
no code implementations • ECCV 2020 • Tong Wang, Yousong Zhu, Chaoyang Zhao, Wei Zeng, Yao-Wei Wang, Jinqiao Wang, Ming Tang
Most of existing object detectors usually adopt a small training batch size ( ~16), which severely hinders the whole community from exploring large-scale datasets due to the extremely long training procedure.
no code implementations • ECCV 2020 • Xin Wen, Biying Li, Haiyun Guo, Zhiwei Liu, Guosheng Hu, Ming Tang, Jinqiao Wang
Some existing methods adopt distribution learning to tackle this issue by exploiting the semantic correlation between age labels.
Ranked #6 on Age Estimation on MORPH album2 (Caucasian)
no code implementations • ECCV 2020 • Xiaomei Zhang, Yingying Chen, Bingke Zhu, Jinqiao Wang, Ming Tang
Although human parsing has made great progress, it still faces a challenge, i. e., how to extract the whole foreground from similar or cluttered scenes effectively.
no code implementations • 2 Apr 2021 • Kuan Zhu, Haiyun Guo, Shiliang Zhang, YaoWei Wang, Gaopan Huang, Honglin Qiao, Jing Liu, Jinqiao Wang, Ming Tang
In this paper, we introduce an alignment scheme in Transformer architecture for the first time and propose the Auto-Aligned Transformer (AAformer) to automatically locate both the human parts and non-human ones at patch-level.
no code implementations • NeurIPS 2021 • Zhaowen Li, Zhiyang Chen, Fan Yang, Wei Li, Yousong Zhu, Chaoyang Zhao, Rui Deng, Liwei Wu, Rui Zhao, Ming Tang, Jinqiao Wang
More importantly, the masked tokens together with the remaining tokens are further recovered by a global image decoder, which preserves the spatial information of the image and is more friendly to the downstream dense prediction tasks.
no code implementations • CVPR 2021 • Linyu Zheng, Ming Tang, Yingying Chen, Guibo Zhu, Jinqiao Wang, Hanqing Lu
Despite considerable similarities between multiple object tracking (MOT) and single object tracking (SOT) tasks, modern MOT methods have not benefited from the development of SOT ones to achieve satisfactory performance.
no code implementations • ICCV 2021 • Bin Yu, Ming Tang, Linyu Zheng, Guibo Zhu, Jinqiao Wang, Hao Feng, Xuetao Feng, Hanqing Lu
End-to-end discriminative trackers improve the state of the art significantly, yet the improvement in robustness and efficiency is restricted by the conventional discriminative model, i. e., least-squares based regression.
no code implementations • 24 Dec 2021 • Zhiwei Liu, Xiangyu Zhu, Lu Yang, Xiang Yan, Ming Tang, Zhen Lei, Guibo Zhu, Xuetao Feng, Yan Wang, Jinqiao Wang
In the second stage, we design a mesh refinement transformer (MRT) to respectively refine each coarse reconstruction result via a self-attention mechanism.
Ranked #65 on 3D Human Pose Estimation on 3DPW (MPJPE metric)
no code implementations • CVPR 2022 • Zhaowen Li, Yousong Zhu, Fan Yang, Wei Li, Chaoyang Zhao, Yingying Chen, Zhiyang Chen, Jiahao Xie, Liwei Wu, Rui Zhao, Ming Tang, Jinqiao Wang
Furthermore, our method can also exploit single-centric-object dataset such as ImageNet and outperforms BYOL by 2. 5% with the same pre-training epochs in linear probing, and surpass current self-supervised object detection methods on COCO dataset, demonstrating its universality and potential.
no code implementations • CVPR 2022 • Tong Wang, Yousong Zhu, Yingying Chen, Chaoyang Zhao, Bin Yu, Jinqiao Wang, Ming Tang
The decision boundary between any two categories is the angular bisector of their weight vectors.
no code implementations • 14 Jun 2022 • Tianyi Yan, Kuan Zhu, Haiyun Guo, Guibo Zhu, Ming Tang, Jinqiao Wang
Clustering-based methods, which alternate between the generation of pseudo labels and the optimization of the feature extraction network, play a dominant role in both unsupervised learning (USL) and unsupervised domain adaptive (UDA) person re-identification (Re-ID).
no code implementations • 31 Aug 2022 • Zhaowen Li, Xu Zhao, Chaoyang Zhao, Ming Tang, Jinqiao Wang
Previous unsupervised domain adaptation methods did not handle the cross-domain problem from the perspective of frequency for computer vision.
no code implementations • 28 Feb 2023 • Zhaowen Li, Yousong Zhu, Zhiyang Chen, Wei Li, Chaoyang Zhao, Liwei Wu, Rui Zhao, Ming Tang, Jinqiao Wang
However, its high random mask ratio would result in two serious problems: 1) the data are not efficiently exploited, which brings inefficient pre-training (\eg, 1600 epochs for MAE $vs.$ 300 epochs for the supervised), and 2) the high uncertainty and inconsistency of the pre-trained model, \ie, the prediction of the same patch may be inconsistent under different mask rounds.
no code implementations • 10 Apr 2023 • Zhaowen Li, Xu Zhao, Peigeng Ding, Zongxin Gao, Yuting Yang, Ming Tang, Jinqiao Wang
In the high-frequency branch, a derivative-filter-like architecture is designed to extract the high-frequency information while a light extractor is employed in the low-frequency branch because the low-frequency information is usually redundant.
no code implementations • 26 Sep 2023 • Qiao Yang, Yu Zhang, Jian Zhang, Zijing Zhao, Shunli Zhang, Jinqiao Wang, Junzhe Chen
Infrared and visible image fusion (IVIF) is used to generate fusion images with comprehensive features of both images, which is beneficial for downstream vision tasks.
no code implementations • 26 Sep 2023 • Qiao Yang, Yu Zhang, Jian Zhang, Zijing Zhao, Shunli Zhang, Jinqiao Wang, Junzhe Chen
Most existing learning-based infrared and visible image fusion (IVIF) methods exhibit massive redundant information in the fusion images, i. e., yielding edge-blurring effect or unrecognizable for object detectors.
no code implementations • 21 Nov 2023 • Zhen Chen, Yuhao Zhai, Jun Zhang, Jinqiao Wang
Specifically, we propose an efficient multi-scale surgical temporal action (MS-STA) module, which integrates visual features with spatial and temporal knowledge of surgical actions at the cost of 2D networks.
no code implementations • 27 Nov 2023 • Jinghan He, Haiyun Guo, Ming Tang, Jinqiao Wang
2) Are the existing three classes of continual learning methods still applicable to the continual instruction tuning of LMMs?
no code implementations • 27 Nov 2023 • Zhiyang Chen, Yousong Zhu, Yufei Zhan, Zhaowen Li, Chaoyang Zhao, Jinqiao Wang, Ming Tang
Large vision-language models (LVLMs) suffer from hallucination a lot, generating responses that apparently contradict to the image content occasionally.
no code implementations • 5 Feb 2024 • Yunfang Niu, Dong Yi, Lingxiang Wu, Zhiwei Liu, Pengxiang Cai, Jinqiao Wang
Virtual try-on can significantly improve the garment shopping experiences in both online and in-store scenarios, attracting broad interest in computer vision.
no code implementations • 29 Feb 2024 • Guojing Ge, Qi Song, Guibo Zhu, Yuting Zhang, Jinglu Chen, Miao Xin, Ming Tang, Jinqiao Wang
Blind face restoration is a challenging task due to the unknown and complex degradation.
no code implementations • 16 Apr 2024 • Enming Zhang, Bingke Zhu, Yingying Chen, Qinghai Miao, Ming Tang, Jinqiao Wang
This limitation restricts the capabilities of pretrained VLMs and can result in incorrect predictions in downstream tasks.