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 Junnan Li

Junnan Li

CoMatch: Semi-supervised Learning with Contrastive Graph Regularization


Nov 23, 2020
Junnan Li, Caiming Xiong, Steven Hoi

Figure 1 for CoMatch: Semi-supervised Learning with Contrastive Graph Regularization
Figure 2 for CoMatch: Semi-supervised Learning with Contrastive Graph Regularization
Figure 3 for CoMatch: Semi-supervised Learning with Contrastive Graph Regularization
Figure 4 for CoMatch: Semi-supervised Learning with Contrastive Graph Regularization

  Access Paper or Ask Questions

MoPro: Webly Supervised Learning with Momentum Prototypes


Sep 17, 2020
Junnan Li, Caiming Xiong, Steven C. H. Hoi

Figure 1 for MoPro: Webly Supervised Learning with Momentum Prototypes
Figure 2 for MoPro: Webly Supervised Learning with Momentum Prototypes
Figure 3 for MoPro: Webly Supervised Learning with Momentum Prototypes
Figure 4 for MoPro: Webly Supervised Learning with Momentum Prototypes

  Access Paper or Ask Questions

The Devil is in Classification: A Simple Framework for Long-tail Instance Segmentation


Jul 31, 2020
Tao Wang, Yu Li, Bingyi Kang, Junnan Li, Junhao Liew, Sheng Tang, Steven Hoi, Jiashi Feng

Figure 1 for The Devil is in Classification: A Simple Framework for Long-tail Instance Segmentation
Figure 2 for The Devil is in Classification: A Simple Framework for Long-tail Instance Segmentation
Figure 3 for The Devil is in Classification: A Simple Framework for Long-tail Instance Segmentation
Figure 4 for The Devil is in Classification: A Simple Framework for Long-tail Instance Segmentation
* LVIS 2019 challenge winner, performance significantly improved after challenge submission, accepted at ECCV 2019 
  Access Paper or Ask Questions

Prototypical Contrastive Learning of Unsupervised Representations


Jun 05, 2020
Junnan Li, Pan Zhou, Caiming Xiong, Richard Socher, Steven C. H. Hoi

Figure 1 for Prototypical Contrastive Learning of Unsupervised Representations
Figure 2 for Prototypical Contrastive Learning of Unsupervised Representations
Figure 3 for Prototypical Contrastive Learning of Unsupervised Representations
Figure 4 for Prototypical Contrastive Learning of Unsupervised Representations

  Access Paper or Ask Questions

Improving out-of-distribution generalization via multi-task self-supervised pretraining


Mar 30, 2020
Isabela Albuquerque, Nikhil Naik, Junnan Li, Nitish Keskar, Richard Socher

Figure 1 for Improving out-of-distribution generalization via multi-task self-supervised pretraining
Figure 2 for Improving out-of-distribution generalization via multi-task self-supervised pretraining
Figure 3 for Improving out-of-distribution generalization via multi-task self-supervised pretraining
Figure 4 for Improving out-of-distribution generalization via multi-task self-supervised pretraining

  Access Paper or Ask Questions

Towards Noise-resistant Object Detection with Noisy Annotations


Mar 03, 2020
Junnan Li, Caiming Xiong, Richard Socher, Steven Hoi

Figure 1 for Towards Noise-resistant Object Detection with Noisy Annotations
Figure 2 for Towards Noise-resistant Object Detection with Noisy Annotations
Figure 3 for Towards Noise-resistant Object Detection with Noisy Annotations
Figure 4 for Towards Noise-resistant Object Detection with Noisy Annotations

  Access Paper or Ask Questions

DivideMix: Learning with Noisy Labels as Semi-supervised Learning


Feb 18, 2020
Junnan Li, Richard Socher, Steven C. H. Hoi

Figure 1 for DivideMix: Learning with Noisy Labels as Semi-supervised Learning
Figure 2 for DivideMix: Learning with Noisy Labels as Semi-supervised Learning
Figure 3 for DivideMix: Learning with Noisy Labels as Semi-supervised Learning
Figure 4 for DivideMix: Learning with Noisy Labels as Semi-supervised Learning
* International Conference on Learning Representations, 2020 
  Access Paper or Ask Questions

Weakly-Supervised Multi-Person Action Recognition in 360$^{\circ}$ Videos


Feb 09, 2020
Junnan Li, Jianquan Liu, Yongkang Wong, Shoji Nishimura, Mohan Kankanhalli

Figure 1 for Weakly-Supervised Multi-Person Action Recognition in 360$^{\circ}$ Videos
Figure 2 for Weakly-Supervised Multi-Person Action Recognition in 360$^{\circ}$ Videos
Figure 3 for Weakly-Supervised Multi-Person Action Recognition in 360$^{\circ}$ Videos
Figure 4 for Weakly-Supervised Multi-Person Action Recognition in 360$^{\circ}$ Videos

  Access Paper or Ask Questions

GradMix: Multi-source Transfer across Domains and Tasks


Feb 09, 2020
Junnan Li, Ziwei Xu, Yongkang Wong, Qi Zhao, Mohan Kankanhalli

Figure 1 for GradMix: Multi-source Transfer across Domains and Tasks
Figure 2 for GradMix: Multi-source Transfer across Domains and Tasks
Figure 3 for GradMix: Multi-source Transfer across Domains and Tasks
Figure 4 for GradMix: Multi-source Transfer across Domains and Tasks

  Access Paper or Ask Questions

Classification Calibration for Long-tail Instance Segmentation


Nov 02, 2019
Tao Wang, Yu Li, Bingyi Kang, Junnan Li, Jun Hao Liew, Sheng Tang, Steven Hoi, Jiashi Feng

Figure 1 for Classification Calibration for Long-tail Instance Segmentation
Figure 2 for Classification Calibration for Long-tail Instance Segmentation
Figure 3 for Classification Calibration for Long-tail Instance Segmentation
Figure 4 for Classification Calibration for Long-tail Instance Segmentation
* This report presents our winning solution to LVIS 2019 challenge 
  Access Paper or Ask Questions

Visual Social Relationship Recognition


Dec 13, 2018
Junnan Li, Yongkang Wong, Qi Zhao, Mohan S. Kankanhalli

Figure 1 for Visual Social Relationship Recognition
Figure 2 for Visual Social Relationship Recognition
Figure 3 for Visual Social Relationship Recognition
Figure 4 for Visual Social Relationship Recognition
* arXiv admin note: text overlap with arXiv:1708.00634 
  Access Paper or Ask Questions

Learning to Learn from Noisy Labeled Data


Dec 13, 2018
Junnan Li, Yongkang Wong, Qi Zhao, Mohan Kankanhalli

Figure 1 for Learning to Learn from Noisy Labeled Data
Figure 2 for Learning to Learn from Noisy Labeled Data
Figure 3 for Learning to Learn from Noisy Labeled Data
Figure 4 for Learning to Learn from Noisy Labeled Data

  Access Paper or Ask Questions

Unsupervised Learning of View-invariant Action Representations


Sep 06, 2018
Junnan Li, Yongkang Wong, Qi Zhao, Mohan S. Kankanhalli

Figure 1 for Unsupervised Learning of View-invariant Action Representations
Figure 2 for Unsupervised Learning of View-invariant Action Representations
Figure 3 for Unsupervised Learning of View-invariant Action Representations
Figure 4 for Unsupervised Learning of View-invariant Action Representations
* NIPS 2018 
  Access Paper or Ask Questions

Interact as You Intend: Intention-Driven Human-Object Interaction Detection


Aug 29, 2018
Bingjie Xu, Junnan Li, Yongkang Wong, Mohan S. Kankanhalli, Qi Zhao

Figure 1 for Interact as You Intend: Intention-Driven Human-Object Interaction Detection
Figure 2 for Interact as You Intend: Intention-Driven Human-Object Interaction Detection
Figure 3 for Interact as You Intend: Intention-Driven Human-Object Interaction Detection
Figure 4 for Interact as You Intend: Intention-Driven Human-Object Interaction Detection

  Access Paper or Ask Questions

Video Storytelling


Jul 25, 2018
Junnan Li, Yongkang Wong, Qi Zhao, Mohan S. Kankanhalli

Figure 1 for Video Storytelling
Figure 2 for Video Storytelling
Figure 3 for Video Storytelling
Figure 4 for Video Storytelling

  Access Paper or Ask Questions

Attention Transfer from Web Images for Video Recognition


Aug 03, 2017
Junnan Li, Yongkang Wong, Qi Zhao, Mohan Kankanhalli

Figure 1 for Attention Transfer from Web Images for Video Recognition
Figure 2 for Attention Transfer from Web Images for Video Recognition
Figure 3 for Attention Transfer from Web Images for Video Recognition
Figure 4 for Attention Transfer from Web Images for Video Recognition
* ACM Multimedia, 2017 
  Access Paper or Ask Questions

Dual-Glance Model for Deciphering Social Relationships


Aug 02, 2017
Junnan Li, Yongkang Wong, Qi Zhao, Mohan S. Kankanhalli

Figure 1 for Dual-Glance Model for Deciphering Social Relationships
Figure 2 for Dual-Glance Model for Deciphering Social Relationships
Figure 3 for Dual-Glance Model for Deciphering Social Relationships
Figure 4 for Dual-Glance Model for Deciphering Social Relationships
* IEEE International Conference on Computer Vision (ICCV), 2017 
  Access Paper or Ask Questions