Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Keze Wang

Solving Inefficiency of Self-supervised Representation Learning


Apr 18, 2021
Guangrun Wang, Keze Wang, Guangcong Wang, Phillip H. S. Torr, Liang Lin

* 11 pages, 3 figures 

  Access Paper or Ask Questions

Temporal Contrastive Graph for Self-supervised Video Representation Learning


Feb 01, 2021
Yang Liu, Keze Wang, Haoyuan Lan, Liang Lin

* 11 pages, 4 figures 

  Access Paper or Ask Questions

Knowledge-Routed Visual Question Reasoning: Challenges for Deep Representation Embedding


Dec 14, 2020
Qingxing Cao, Bailin Li, Xiaodan Liang, Keze Wang, Liang Lin

* To appear in TNNLS 2021. Considering that a desirable VQA model should correctly perceive the image context, understand the question, and incorporate its learned knowledge, our proposed dataset aims to cutoff the shortcut learning exploited by the current deep embedding models and push the research boundary of the knowledge-based visual question reasoning 

  Access Paper or Ask Questions

Continuous Transition: Improving Sample Efficiency for Continuous Control Problems via MixUp


Nov 30, 2020
Junfan Lin, Zhongzhan Huang, Keze Wang, Xiaodan Liang, Weiwei Chen, Liang Lin


  Access Paper or Ask Questions

Linguistically Driven Graph Capsule Network for Visual Question Reasoning


Mar 23, 2020
Qingxing Cao, Xiaodan Liang, Keze Wang, Liang Lin

* Submitted to TPAMI 2020. We have achieved an end-to-end interpretable structural reasoning for general images without the requirement of layout annotations 

  Access Paper or Ask Questions

Learning Reinforced Agents with Counterfactual Simulation for Medical Automatic Diagnosis


Mar 14, 2020
Junfan Lin, Ziliang Chen, Xiaodan Liang, Keze Wang, Liang Lin

* Submitted to TPAMI 2020. In the experiments, our trained agent achieves the new state-of-the-art under various experimental settings and possesses the advantage of sample-efficiency and robustness compared to other existing MAD methods 

  Access Paper or Ask Questions

Instance-Aware Representation Learning and Association for Online Multi-Person Tracking


May 29, 2019
Hefeng Wu, Yafei Hu, Keze Wang, Hanhui Li, Lin Nie, Hui Cheng

* accepted by Pattern Recognition 

  Access Paper or Ask Questions

Face Hallucination by Attentive Sequence Optimization with Reinforcement Learning


May 04, 2019
Yukai Shi, Guanbin Li, Qingxing Cao, Keze Wang, Liang Lin

* To be published in TPAMI 

  Access Paper or Ask Questions

Adaptively Connected Neural Networks


Apr 07, 2019
Guangrun Wang, Keze Wang, Liang Lin

* Accepted by CVPR 2019 

  Access Paper or Ask Questions

3D Human Pose Machines with Self-supervised Learning


Jan 15, 2019
Keze Wang, Liang Lin, Chenhan Jiang, Chen Qian, Pengxu Wei

* To appear in IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI), 2019. Our simple yet effective self-supervised correction mechanism to incorporate 3D pose geometric structural information is innovative in the literature, and may also inspire other 3D vision tasks. Please find the code of this project at: http://www.sysu-hcp.net/3d_pose_ssl/ 

  Access Paper or Ask Questions

Cost-effective Object Detection: Active Sample Mining with Switchable Selection Criteria


Sep 16, 2018
Keze Wang, Liang Lin, Xiaopeng Yan, Ziliang Chen, Dongyu Zhang, Lei Zhang

* Automatically determining whether an unlabeled sample should be manually annotated or pseudo-labeled via a novel self-learning process (Accepted by TNNLS 2018) The source code is available at http://kezewang.com/codes/ASM_ver1.zip 

  Access Paper or Ask Questions

Towards Human-Machine Cooperation: Self-supervised Sample Mining for Object Detection


May 24, 2018
Keze Wang, Xiaopeng Yan, Dongyu Zhang, Lei Zhang, Liang Lin

* We enabled to mine from unlabeled or partially labeled data to boost object detection (Accepted by CVPR 2018) The source code is available at http://kezewang.com/codes/SSM_CVPR.zip 

  Access Paper or Ask Questions

Recurrent 3D Pose Sequence Machines


Jul 31, 2017
Mude Lin, Liang Lin, Xiaodan Liang, Keze Wang, Hui Cheng

* Published in CVPR 2017 

  Access Paper or Ask Questions

Deep Co-Space: Sample Mining Across Feature Transformation for Semi-Supervised Learning


Jul 28, 2017
Ziliang Chen, Keze Wang, Xiao Wang, Pai Peng, Ebroul Izquierdo, Liang Lin

* To appear in IEEE Transactions on Circuits and Systems for Video Technology (T-CSVT), 2017 

  Access Paper or Ask Questions

Structure-Preserving Image Super-resolution via Contextualized Multi-task Learning


Jul 26, 2017
Yukai Shi, Keze Wang, Chongyu Chen, Li Xu, Liang Lin

* To appear in Transactions on Multimedia 2017 

  Access Paper or Ask Questions

Active Self-Paced Learning for Cost-Effective and Progressive Face Identification


Jul 03, 2017
Liang Lin, Keze Wang, Deyu Meng, Wangmeng Zuo, Lei Zhang

* To appear in IEEE Transactions on Pattern Analysis and Machine Intelligence 2017 

  Access Paper or Ask Questions

Cost-Effective Active Learning for Deep Image Classification


Jan 13, 2017
Keze Wang, Dongyu Zhang, Ya Li, Ruimao Zhang, Liang Lin

* Accepted by IEEE Transactions on Circuits and Systems for Video Technology (TCSVT) 2016 

  Access Paper or Ask Questions

Human Pose Estimation from Depth Images via Inference Embedded Multi-task Learning


Aug 13, 2016
Keze Wang, Shengfu Zhai, Hui Cheng, Xiaodan Liang, Liang Lin

* To appear in ACM Multimedia 2016, full paper (oral), 10 pages, 11 figures 

  Access Paper or Ask Questions

Local- and Holistic- Structure Preserving Image Super Resolution via Deep Joint Component Learning


Jul 25, 2016
Yukai Shi, Keze Wang, Li Xu, Liang Lin

* Published on ICME 2016 (oral), 6 pages, 6 figures 

  Access Paper or Ask Questions

A Deep Structured Model with Radius-Margin Bound for 3D Human Activity Recognition


Dec 05, 2015
Liang Lin, Keze Wang, Wangmeng Zuo, Meng Wang, Jiebo Luo, Lei Zhang

* International Journal of Computer Vision, Volume 118, Issue 2, pp 256-273 (June 2016) 
* 16 pages, 9 figures, to appear in International Journal of Computer Vision 2015 

  Access Paper or Ask Questions

PISA: Pixelwise Image Saliency by Aggregating Complementary Appearance Contrast Measures with Edge-Preserving Coherence


May 13, 2015
Keze Wang, Liang Lin, Jiangbo Lu, Chenglong Li, Keyang Shi

* IEEE Transactions on Image Processing (TIP), volume. 24, Issue. 10, page. 3019 - 3033, Oct. 2015 
* 14 pages, 14 figures, 1 table, to appear in IEEE Transactions on Image Processing 

  Access Paper or Ask Questions

3D Human Activity Recognition with Reconfigurable Convolutional Neural Networks


Feb 01, 2015
Keze Wang, Xiaolong Wang, Liang Lin, Meng Wang, Wangmeng Zuo

* This manuscript has 10 pages with 9 figures, and a preliminary version was published in ACM MM'14 conference 

  Access Paper or Ask Questions