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 Dahua Lin

Towards Evaluating and Training Verifiably Robust Neural Networks


Apr 05, 2021
Zhaoyang Lyu, Minghao Guo, Tong Wu, Guodong Xu, Kehuan Zhang, Dahua Lin

* Accepted to CVPR 2021 (Oral) 

  Access Paper or Ask Questions

CARAFE++: Unified Content-Aware ReAssembly of FEatures


Dec 07, 2020
Jiaqi Wang, Kai Chen, Rui Xu, Ziwei Liu, Chen Change Loy, Dahua Lin

* Technical Report. Extended journal version of the conference paper that appeared as arXiv:1905.02188 

  Access Paper or Ask Questions

FLAVA: Find, Localize, Adjust and Verify to Annotate LiDAR-Based Point Clouds


Nov 20, 2020
Tai Wang, Conghui He, Zhe Wang, Jianping Shi, Dahua Lin

* Full technical report for the UIST 2020 Poster version 

  Access Paper or Ask Questions

Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation


Nov 19, 2020
Xinge Zhu, Hui Zhou, Tai Wang, Fangzhou Hong, Yuexin Ma, Wei Li, Hongsheng Li, Dahua Lin

* This work achieves the 1st place in the leaderboard of SemanticKITTI (until CVPR DDL) and based on this work, we also achieve the 1st place in the leaderboard of SemanticKITTI panoptic segmentation; Code at https://github.com/xinge008/Cylinder3D 

  Access Paper or Ask Questions

Understanding the wiring evolution in differentiable neural architecture search


Sep 07, 2020
Sirui Xie, Shoukang Hu, Xinjiang Wang, Chunxiao Liu, Jianping Shi, Xunying Liu, Dahua Lin


  Access Paper or Ask Questions

Seesaw Loss for Long-Tailed Instance Segmentation


Aug 23, 2020
Jiaqi Wang, Wenwei Zhang, Yuhang Zang, Yuhang Cao, Jiangmiao Pang, Tao Gong, Kai Chen, Ziwei Liu, Chen Change Loy, Dahua Lin

* Technical Report for LVIS Challenge 2020 https://www.lvisdataset.org/challenge_2020 

  Access Paper or Ask Questions

A Unified Framework for Shot Type Classification Based on Subject Centric Lens


Aug 08, 2020
Anyi Rao, Jiaze Wang, Linning Xu, Xuekun Jiang, Qingqiu Huang, Bolei Zhou, Dahua Lin

* ECCV2020. Project page: https://anyirao.com/projects/ShotType.html 

  Access Paper or Ask Questions

Online Multi-modal Person Search in Videos


Aug 08, 2020
Jiangyue Xia, Anyi Rao, Qingqiu Huang, Linning Xu, Jiangtao Wen, Dahua Lin

* ECCV2020. Project page: http://movienet.site/projects/eccv20onlineperson.html 

  Access Paper or Ask Questions

Cylinder3D: An Effective 3D Framework for Driving-scene LiDAR Semantic Segmentation


Aug 04, 2020
Hui Zhou, Xinge Zhu, Xiao Song, Yuexin Ma, Zhe Wang, Hongsheng Li, Dahua Lin

* Source code: https://github.com/xinge008/Cylinder3D 

  Access Paper or Ask Questions

Distribution-Balanced Loss for Multi-Label Classification in Long-Tailed Datasets


Jul 25, 2020
Tong Wu, Qingqiu Huang, Ziwei Liu, Yu Wang, Dahua Lin

* To appear in ECCV 2020 as a spotlight presentation. Code and models are available at: https://github.com/wutong16/DistributionBalancedLoss 

  Access Paper or Ask Questions

MovieNet: A Holistic Dataset for Movie Understanding


Jul 21, 2020
Qingqiu Huang, Yu Xiong, Anyi Rao, Jiaze Wang, Dahua Lin

* Accepted by ECCV2020 as spotlight presentation. Project page: http://movienet.site 

  Access Paper or Ask Questions

Sep-Stereo: Visually Guided Stereophonic Audio Generation by Associating Source Separation


Jul 20, 2020
Hang Zhou, Xudong Xu, Dahua Lin, Xiaogang Wang, Ziwei Liu

* To appear in Proceedings of the European Conference on Computer Vision (ECCV), 2020. Code, models, and video results are available on our webpage: https://hangz-nju-cuhk.github.io/projects/Sep-Stereo 

  Access Paper or Ask Questions

Placepedia: Comprehensive Place Understanding with Multi-Faceted Annotations


Jul 17, 2020
Huaiyi Huang, Yuqi Zhang, Qingqiu Huang, Zhengkui Guo, Ziwei Liu, Dahua Lin

* To appear in ECCV 2020. Dataset is available at: https://hahehi.github.io/placepedia.html 

  Access Paper or Ask Questions

Learn to Propagate Reliably on Noisy Affinity Graphs


Jul 17, 2020
Lei Yang, Qingqiu Huang, Huaiyi Huang, Linning Xu, Dahua Lin

* 14 pages, 7 figures, ECCV 2020 

  Access Paper or Ask Questions

Novel Policy Seeking with Constrained Optimization


May 21, 2020
Hao Sun, Zhenghao Peng, Bo Dai, Jian Guo, Dahua Lin, Bolei Zhou


  Access Paper or Ask Questions

Intra- and Inter-Action Understanding via Temporal Action Parsing


May 20, 2020
Dian Shao, Yue Zhao, Bo Dai, Dahua Lin

* CVPR 2020 Poster; Project page: https://sdolivia.github.io/TAPOS/ 

  Access Paper or Ask Questions

Evolutionary Stochastic Policy Distillation


Apr 30, 2020
Hao Sun, Xinyu Pan, Bo Dai, Dahua Lin, Bolei Zhou


  Access Paper or Ask Questions

Motion Guided 3D Pose Estimation from Videos


Apr 29, 2020
Jingbo Wang, Sijie Yan, Yuanjun Xiong, Dahua Lin


  Access Paper or Ask Questions

A Local-to-Global Approach to Multi-modal Movie Scene Segmentation


Apr 28, 2020
Anyi Rao, Linning Xu, Yu Xiong, Guodong Xu, Qingqiu Huang, Bolei Zhou, Dahua Lin

* CVPR2020. Project page: https://anyirao.com/projects/SceneSeg.html 

  Access Paper or Ask Questions

FineGym: A Hierarchical Video Dataset for Fine-grained Action Understanding


Apr 14, 2020
Dian Shao, Yue Zhao, Bo Dai, Dahua Lin

* CVPR 2020 Oral (3 strong accepts); Project page: https://sdolivia.github.io/FineGym/ 

  Access Paper or Ask Questions

Feature Pyramid Grids


Apr 07, 2020
Kai Chen, Yuhang Cao, Chen Change Loy, Dahua Lin, Christoph Feichtenhofer

* Technical report 

  Access Paper or Ask Questions

Self-Supervised Scene De-occlusion


Apr 06, 2020
Xiaohang Zhan, Xingang Pan, Bo Dai, Ziwei Liu, Dahua Lin, Chen Change Loy

* Accepted to CVPR 2020 as oral. Project page: https://xiaohangzhan.github.io/projects/deocclusion/ 

  Access Paper or Ask Questions

SSN: Shape Signature Networks for Multi-class Object Detection from Point Clouds


Apr 06, 2020
Xinge Zhu, Yuexin Ma, Tai Wang, Yan Xu, Jianping Shi, Dahua Lin

* Code is available at https://github.com/xinge008/SSN 

  Access Paper or Ask Questions

Reconfigurable Voxels: A New Representation for LiDAR-Based Point Clouds


Apr 06, 2020
Tai Wang, Xinge Zhu, Dahua Lin


  Access Paper or Ask Questions