Wenhai Wang

About Me ([GitHub] [Google Scholar] [CV])

I am a Research Scientist at Shanghai AI Laboratory, led by Prof. Jifeng Dai and Prof. Yu Qiao

Previously, I obtained the Ph.D. degree from Department of Computer Science and Technology, Nanjing University (NJU) in 2021. My academic supervisor is Prof. Tong Lu. I received my B.E degree from Nanjing University of Science and Technology (NUST) in 2016.
I work very close with my friends Dr. Enze Xie and Prof. Xiang Li. I was fortunate to work with Prof. Ping Luo and Prof. Chunhua Shen.

My recent works are mainly on:

CNN/Transformer Backbone
Object Detection & Semantic/Instance/Panoptic Segmentation
Vision-Language Model
Autonomous Driving Perception
Optical Character Recognition

The fundamental vision department at Shanghai AI Laboratory is now hiring. If you are interested in internship/researcher positions related to computer vision, please feel free to contact me through the email.

News

2022/07/04: Two papers are accepted by ECCV 2022. See BEVFormer，VL-LTR.
2022/06/30: EHNet is accepted by ACM MM 2022.
2022/06/20: Our team wins the champion of Waymo 2022 3D Camera-Only Detection Task (15,000 USD Bonus).
2022/05/25: An extended version of GFocalv1/v2 is accepted by TPAMI 2022.
2022/04/26: I am selected as one of TechBeat 2022 Most Popular Speakers.
2022/03/10: BEVFormer ranks the 1st on nuScenes camera-only 3D detection benchmark.
2022/03/01: Panoptic SegFormer is accepted by CVPR 2022.
2022/02/08: PVT v2 is accepted by CVMJ 2022.
2022/02/03: PVT is selected as one of ICCV21' Top-10 Influential Papers (Rank 2), and SegFormer is selected as one of NeurIPS21' Top-10 Influential Papers (Rank 3).
2021/12/01: URST is accepted by AAAI 2022.
2021/09/30: SegFormer is accepted by NeurIPS 2021.
2021/08/27: I defend my Ph.D. dissertation, and became a Ph.D.
2021/07/23: Two papers are accepted by ICCV 2021. See PVT (oral), DetCo.
2021/05/05: Two papers are accepted by TPAMI 2021. See PAN++, PolarMask++.
2021/04/29: Trans2Seg is accepted by IJCAI 2021.
2021/04/01: We rewrite the code of PSENet, making it clear and easier to use. See here.
2021/03/01: GFocalv2 is accepted by CVPR 2021.
2021/02/24: Code of PVT is released at here.
2021/02/11: PolarMask is selected as one of CVPR20' Top-10 Influential Papers.
2020/12/21: Our team wins the champion of NAIC 2020 Remote Sensing Semantic Segmentation Task (1,000,000 RMB bonus).
2020/10/13: Code of AE TextSpotter is released at here.
2020/09/25: GFocal is accepted by NeurIPS 2020.
2020/09/02: Code of PAN is released at here.
2020/07/03: Four papers are accepted by ECCV 2020. See AE TextSpotter, TransLab, HGG, TSRN.
2020/03/10: PolarMask (oral) is accepted by CVPR 2020.
2019/07/23: PAN is accepted by ICCV 2019.
2019/03/15: Two paper are accepted by CVPR 2019. See PSENet, SKNet.
2018/06/16: MixNet (oral) is accepted by IJCAI 2018.

Experience

2021/09 - Present: Research Scientist at Shanghai AI Laboratory, led by Prof. Jifeng Dai and Prof. Yu Qiao
2019/10 - 2020/03: Research Assistant at the University of Hong Kong (HKU), led by Prof. Ping Luo
2019/08 - 2020/03: Research Intern at SenseTime Group Limited, led by Xuebo Liu and Ding Liang
2018/06 - 2018/12: Research Intern at Momenta, led by Prof. Xiang Li

Recent Works ([Full List])

(* Equal contribution, † Interns, # Corresponding authors)

Vision Transformer Adapter for Dense Predictions
Zhe Chen*†, Yuchen Duan*†, Wenhai Wang*, Junjun He, Tong Lu, Jifeng Dai, Yu Qiao#
arXiv, 2022
[Paper] [Code] Star [BibTex]
We design a ViT adapter for dense prediction tasks.

BEVFormer: Learning Bird’s-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers
Zhiqi Li*†, Wenhai Wang*, Hongyang Li*, Enze Xie, Chonghao Sima, Tong Lu, Yu Qiao, Jifeng Dai#
ECCV, 2022
[Paper] [Code] Star [BibTex]
A versatile camera-only framework for autonomous driving perception, e.g., 3D object detection and semantic map segmentation.

VL-LTR: Learning Class-wise Visual-Linguistic Representation for Long-Tailed Visual Recognition
Changyao Tian*†, Wenhai Wang*, Xizhou Zhu*, Xiaogang Wang, Jifeng Dai#, Yu Qiao
ECCV, 2022
[Paper] [Code] Star [BibTex]
We design a vision-language-based framework for long-tailed recognition.

Generalized Focal Loss: Towards Efficient Representation Learning for Dense Object Detection
Xiang Li, Chengqi Lv, Wenhai Wang, Gang Li, Lingfeng Yang, Jian Yang#
TPAMI, 2022
[Paper] [Code] Star [BibTex]
An extended version of GFLv1/v2.

Panoptic SegFormer: Delving Deeper into Panoptic Segmentation with Transformers
Zhiqi Li†, Wenhai Wang#, Enze Xie, Zhiding Yu, Anima Anandkumar, Jose M. Alvarez, Tong Lu#, Ping Luo
CVPR, 2022
[Paper] [Code] Star [BibTex]
We proposed a transformer-based panoptic segmentation framework.

PVT v2: Improved Baselines with Pyramid Vision Transformer
Wenhai Wang#, Enze Xie, Xiang Li, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping Luo, Ling Shao
CVMJ, 2021
[Paper] [Code] Star [中文解读] [Report] [Talk] [BibTex]
A better PVT.

Towards Ultra-Resolution Neural Style Transfer via Thumbnail Instance Normalization
Zhe Chen†, Wenhai Wang#, Enze Xie, Tong Lu#, Ping Luo
AAAI, 2022
[Paper] [Code] Star [BibTex]
We proposed a neural style transfer framework for arbitrary ultra-resolution images.

Selected Works ([Full List])

Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions
Wenhai Wang, Enze Xie, Xiang Li, Deng-Ping Fan#, Kaitao Song, Ding Liang, Tong Lu#, Ping Luo, Ling Shao
ICCV, 2021 (oral presentation)
[Paper] [Code] Star [中译版] [中文解读] [Report] [Talk] [BibTex]
[ICCV21' Top-10 Influential Papers]
A pure Transformer backbone for dense prediction, such as object detection and semantic segmentation.

PAN++: Towards Efficient and Accurate End-to-End Spotting of Arbitrarily-Shaped Text
Wenhai Wang*, Enze Xie*, Xiang Li, Xuebo Liu, Ding Liang, Zhibo Yang, Tong Lu#, Chunhua Shen
TPAMI, 2021
[Paper] [Code] Star [BibTex]
We extend PSENet (CVPR'19) and PAN (ICCV'19) to a text spotting system.

Shape Robust Text Detection with Progressive Scale Expansion Network
Wenhai Wang*, Enze Xie*, Xiang Li, Wenbo Hou, Tong Lu#, Gang Yu, Shuai Shao
CVPR, 2019
[Paper] [Poster] [Code] Star [BibTex]
We proposed a segmentation-based text detector that can precisely detect text instances with arbitrary shapes.

PolarMask++: Enhanced Polar Representation for Single-Shot Instance Segmentation and Beyond
Enze Xie*, Wenhai Wang*, Mingyu Ding, Ruimao Zhang, Ping Luo#
TPAMI, 2021
[Paper] [Code] Star [BibTex]
[CVPR20' Top-10 Influential Papers]
We extend PolarMask(CVPR'20) to several instance-level detection tasks.

SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers
Enze Xie, Wenhai Wang, Zhiding Yu#, Anima Anandkuma, Jose M. Alvarez, Ping Luo#
NeurIPS, 2021
[Paper] [Code] Star [中文解读] [Demo] [BibTex]
[NeurIPS21' Top-10 Influential Papers]
A simple and effective Transformer-based semantic segmentation framework.

Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detection
Xiang Li, Wenhai Wang, Lijun Wu, Shuo Chen, Xiaolin Hu, Jun Li, Jinhui Tang, Jian Yang#
NeurIPS, 2020
[Paper] [Code] Star [BibTex]
We propose the generalized focal loss for learning the improved representations of dense object detector.

Selective Kernel Networks
Xiang Li, Wenhai Wang, Xiaolin Hu, Jian Yang#
CVPR, 2019
[Paper] [Code] Star [BibTex]
We proposed a dynamic selection mechanism in convolutional neural networks.

Honors and Awards

2022/06: Waymo 2022 3D Camera-Only Detection Task, 1st Place (15,000 USD Bonus)
2020/12: National Artificial Intelligence Challenge (NAIC) 2020, Remote Sensing Semantic Segmentation Task, 1st Place (1,000,000 RMB Bonus)
2019/12: China National Scholarship
2019/09: ICDAR2019 Robust Reading Challenge on Arbitrary-Shaped Text, Task1, 1st Place
2019/09: ICDAR2019 Robust Reading Challenge on Large-scale Street View Text with Partial Labeling, Task1, 2nd Place
2018/12: AI Challenger 2018 Autonomous Driving Perception Task, 2nd Place (40,000 RMB Bonus)
2015/11: ACM-ICPC Asia Regional Contest, Silver Medal

Review Services

Senior Program Committee Member
International Joint Conference on Artificial Intelligence (IJCAI), 2021

Journal Reviewer
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
International Journal of Computer Vision (IJCV)
IEEE Transactions on Image Processing (TIP)
IEEE Transactions on Multimedia (TMM)
Computational Visual Media Journal (CVMJ)
Pattern Recognition (PR)

Program Committee Member/Conference Reviewer
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020, 2021, 2022
Neural Information Processing Systems (NeurIPS), 2020, 2021
International Conference on Machine Learning (ICML), 2021, 2022
International Conference on Learning Representations (ICLR), 2021
IEEE International Conference on Computer Vision (ICCV), 2021
European Conference on Computer Vision (ECCV), 2022
AAAI Conference on Artificial Intelligence (AAAI), 2022
International Joint Conference on Artificial Intelligence (IJCAI), 2022
IEEE Winter Conference on Applications of Computer Vision (WACV), 2021
Asian Conference on Computer Vision (ACCV), 2020