Publications

Semi-supervised Domain Adaptation based on Dual-level Domain Mixing for Semantic Segmentation

Published in CVPR, 2021

A novel framework based on dual-level domain mixing is proposed. The proposed framework consists of three stages. First, two kinds of data mixing methods are proposed to reduce domain gap in both region-level and sample-level respectively. We can obtain two complementary domain-mixed teachers based on dual-level mixed data from holistic and partial views respectively.

Recommended citation: ShuaiJun Chen, Xu Jia, Jianzhong He and etal. (2021). "Semi-supervised Domain Adaptation based on Dual-level Domain Mixing for Semantic Segmentation." CVPR 2021. https://arxiv.org/abs/2103.04705

Can Semantic Labels Assist Self-Supervised Visual Representation Learning?

Published in Arxiv, 2021

In this paper, we defend the usefulness of semantic labels but point out that fully-supervised and self-supervised methods are pursuing different kinds of features. To alleviate this issue, we present a new algorithm named Supervised Contrastive Adjustment in Neighborhood (SCAN) that maximally prevents the semantic guidance from damaging the appearance feature embedding.

Recommended citation: Longhui Wei, Lingxi Xie, Jianzhong He and etal. (2021). "Can Semantic Labels Assist Self-Supervised Visual Representation Learning?." Arxiv. https://arxiv.org/abs/2011.08621

Multi-Target Domain Adaptation With Collaborative Consistency Learning

Published in CVPR, 2021

This paper is about domain adaptation for semantic segmentation

Recommended citation: Takashi Isobe, Xu Jia, Shuajun Chen, Jianzhong He, and etal. (2021). "Multi-Target Domain Adaptation With Collaborative Consistency Learning." CVPR 2021. https://arxiv.org/abs/2103.04717

Multi-Source Domain Adaptation with Collaborative Learning for Semantic Segmentation

Published in CVPR, 2021

In this paper, we propose a novel multi-source domain adaptation framework based on collaborative learning for semantic segmentation.

Recommended citation: Jianzhong He, Xu Jia and etal. (2021). "Multi-Source Domain Adaptation with Collaborative Learning for Semantic Segmentation." CVPR 2021. https://arxiv.org/abs/2103.04717

ATSO: Asynchronous Teacher-Student Optimizationfor Semi-Supervised Medical Image Segmentation

Published in CVPR, 2021

This paper is about the semi- and self- supervised segmentation of medical images and the auto driving scene.

Recommended citation: Xinyue Huo, Lingxi Xie, Jianzhong He, and etal. (2020). "ATSO: Asynchronous Teacher-Student Optimizationfor Semi-Supervised Medical Image Segmentation." CVPR 2021. http://pkuCactus.github.io/files/ATSO.pdf

BDCN: Bi-Directional Cascade Network for Perceptual Edge Detection

Published in IEEE T-PAMI, 2020

This paper is about edge detection, which to explore the multi-scale information of images, and adopt the detected edges to help other high level vision tasks

Recommended citation: Jianzhong He, Shiliang Zhang and etal. (2020). "BDCN: Bi-Directional Cascade Network for Perceptual Edge Detection." IEEE T-PAMI 2020. http://pkuCactus.github.io/files/BDCN-PAMI.pdf

Bi-Directional Cascade Network for Perceptual Edge Detection

Published in CVPR, 2019

This paper is about the edge detection of natural images, and we aplly a cascade method to solve it

Recommended citation: Jianzhong He, Shiliang Zhang and etal. (2019). "Bi-Directional Cascade Network for Perceptual Edge Detection." CVPR 2019. http://pkuCactus.github.io/files/BDCN-CVPR.pdf

EAGER: Edge-Aided imaGe undERstanding System

Published in ICMR, 2019

This paper is edge detection and the help of which to the optical flow, segmentation and proposal.

Recommended citation: Jianzhong He and etal. (2019). "EAGER: Edge-Aided imaGe undERstanding System" ICMR 2019 http://pkuCactus.github.io/files/EAGER.pdf