EVA: Exploring the Limits of Masked Visual Representation Learning at Scale

IEEE Conference on Computer Vision and Pattern Recognition (CVPR Highlights), 2023

PaperDigest Most Influential Papers

Yuxin Fang, Wen Wang, Binhui Xie, Quan Sun, Ledell Yu Wu, Xinggang Wang, Tiejun Huang, Xinlong Wang, and Yue Cao

EVA is the first open-sourced billion-scale vision foundation model that achieves state-of-the-art performance on a broad range of downstream tasks.

SePiCo: Semantic-Guided Pixel Contrast for Domain Adaptive Semantic Segmentation

IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI, IF: 23.6), 2023

ESI Highly Cited Paper

Binhui Xie, Shuang Li, Mingjia Li, Chi Harold Liu, Gao Huang, and Guoren Wang

A novel one-stage adaptation framework that highlights the semantic concepts of individual pixel to promote learning of class-discriminative and class-balanced pixel embedding space across domains.

Annotator: A Generic Active Learning Baseline for LiDAR Semantic Segmentation

Advances in Neural Information Processing Systems (NeurIPS), 2023

Binhui Xie, Shuang Li, Qingju Guo, Chi Harold Liu, and Xinjing Cheng

We established a simple and general baseline for label-efficient LiDAR semantic segmentation.

Towards Fewer Annotations: Active Learning via Region Impurity and Prediction Uncertainty for Domain Adaptive Semantic Segmentation

IEEE Conference on Computer Vision and Pattern Recognition (CVPR Oral), 2022

Binhui Xie, Longhui Yuan, Shuang Li, Chi Harold Liu, and Xinjing Cheng

A region-based acquisition strategy for DA Segmentation queries image regions that are both diverse in spatial adjacency and uncertain in prediction output.

Active Learning for Domain Adaptation: An Energy-based Approach

AAAI Conference on Artificial Intelligence (AAAI), 2022

Binhui Xie, Longhui Yuan, Shuang Li, Chi Harold Liu, Xinjing Cheng, and Guoren Wang

A new perspective to select a highly informative subset of unlabeled target data under domain shift via exploiting free energy biases across domains.

A Collaborative Alignment Framework of Transferable Knowledge Extraction for Unsupervised Domain Adaptation

IEEE Transactions on Knowledge and Data Engineering (TKDE), 2023

ESI Highly Cited Paper

Binhui Xie, Shuang Li, Fangrui Lv, Chi Harold Liu, Guoren Wang, and Dapeng Wu

We introduce a collaborative alignment framework that integrates global structure information and local semantic consistency into a unified deep model.

Generalized Domain Conditioned Adaptation Network

IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI, IF: 23.6), 2022

Shuang Li, Binhui Xie, Qiuxia Lin, Chi Harold Liu, Gao Huang, and Guoren Wang

We develop GDCAN to automatically determine whether domain channel activations should be separately modeled in each attention module for domain adaptaion problem.

Robust Test-Time Adaptation in Dynamic Scenarios

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023

Longhui Yuan, Binhui Xie and Shuang Li

RoTTA presents a new practical test-time adaptation setting where environments gradually change and the test data is sampled correlatively over time.

VBLC: Visibility Boosting and Logit-Constraint Learning for Domain Adaptive Semantic Segmentation under Adverse Conditions

AAAI Conference on Artificial Intelligence (AAAI Oral), 2023

Mingjia Li*, Binhui Xie*, Shuang Li, Chi Harold Liu, and Xinjing Cheng

VBLC tackles the problem of domain adaptive semantic segmentation under adverse conditions, getting rid of reference images.

Simultaneous Semantic Alignment Network for Heterogeneous Domain Adaptation

ACM International Conference on Multimedia (ACM MM), 2020

Shuang Li, Binhui Xie, Jiashu Wu, Ying Zhao, Chi Harold Liu, and Zhengming Ding

A HDA method simultaneously exploits correlations among categories and aligns the centroids for each category across domains.