Research

Research Interest

My research interest focuses on deep learning with applications in computer vision and medical imaging. Specific topics I've been working on are as follows:

・Label efficient learning: 

☑︎ weakly supervised learning, e.g. scribbles, extreme points, RECIST labels

☑︎ semi-supervised learning

☑︎ self-supervised learning, contrastive learning

・Continual learning: 

☑︎ class incremental learning, w/o memory or exemplar

☑︎ continual semantic segmentation

・Multi-modality learning: 

☑︎ vision-language representation learning

☑︎ image captioning, text-to-image generation

☑︎ multi-modality medical segmentation, e.g. multi-contrast multi-channel CT/MRI

・Transfer learning: 

☑︎ domain adaptation

☑︎ knowledge distillation

・Real-world applications in vision system: 

☑︎ interactive segmentation

☑︎ 3D segmentation, e.g. multi-organ segmentation, tumor segmentation

☑︎ image-text matching/retrieval

Publications

Conference Papers

• Continual Domain Adversarial Adaptation via Double-Head Discriminators  AISTATS 

Yan Shen, Zhanghexuan Ji, Chunwei Ma, Mingchen Gao. International Conference on Artificial Intelligence and Statistics (AISTATS), 2024. [Arxiv] [BibTex]

• Continual Segment: Towards a Single, Unified and Accessible Continual Segmentation Model of 143 Whole-body Organs in CT Scans  ICCV 

Zhanghexuan Ji*, Dazhou Guo*, Puyang Wang, Ke Yan, Jia Ge, Xianghua Ye, Minfeng Xu, Jingren Zhou, Le Lu, Mingchen Gao, Dakai Jin. International Conference on Computer Vision (ICCV), 2023. [ICCV] [Arxiv] [BibTex]

• Progressive Voronoi Diagram Subdivision Enables Accurate Data-free Class-Incremental Learning  ICLR 

Chunwei Ma, Zhanghexuan Ji, Ziyun Huang, Yan Shen, Mingchen Gao, Jinhui Xu. International Conference on Learning Representations (ICLR), 2023. [OpenReview] [Arxiv] [BibTex]

• FedMM: A Communication Efficient Solver for Federated Adversarial Domain Adaptation  AAMAS 

Yan Shen, Jian Du, Han Zhao, Zhanghexuan Ji, Chunwei Ma and Mingchen Gao. International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2023. [PDF] [Arxiv] [BibTex]

• Improving Joint Learning of Chest X-Ray and Radiology Report by Word Region Alignment  MLMI 

Zhanghexuan Ji*, Mohammad Abuzar Shaikh*, Dana Moukheiber, Sargur N Srihari, Yifan Peng, Mingchen Gao. International Workshop on Machine Learning in Medical Imaging (MICCAIW-MLMI), 2021. [Arxiv] [Code] [BibTex]

• User-Guided Domain Adaptation for Rapid Annotation from User Interactions: A Study on Pathological Liver Segmentation  MICCAI 

Ashwin Raju, Zhanghexuan Ji, Chi Tung Cheng, Jinzheng Cai, Junzhou Huang, Jing Xiao, Le Lu, ChienHung Liao, Adam P Harrison. International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2020. [Arxiv] [BibTex]

• An End-to-End Learnable Flow Regularized Model for Brain Tumor Segmentation  MLMI 

Yan Shen, Zhanghexuan Ji, Mingchen Gao. International Workshop on Machine Learning in Medical Imaging (MICCAIW-MLMI), 2020. [Arxiv] [BibTex]

• Scribble-based Hierarchical Weakly Supervised Learning for Brain Tumor Segmentation  MICCAI 

Zhanghexuan Ji, Yan Shen, Chunwei Ma, Mingchen Gao. International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2019. [Arxiv] [BibTex]

• Neural Style Transfer Improves 3D Cardiovascular MR Image Segmentation on Inconsistent Data  MICCAI 

Chunwei Ma, Zhanghexuan Ji, Mingchen Gao. International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2019. [Arxiv] [BibTex]

Preprints

• LAViTeR: Learning Aligned Visual and Textual Representations Assisted by Image and Caption Generation

Mohammad Abuzar Shaikh*, Zhanghexuan Ji*, Dana Moukheiber, Yan Shen, Sargur Srihari, Mingchen Gao. Preprint, 2021. [Arxiv] [BibTex]

• A Bayesian Detect to Track System for Robust Visual Object Tracking and Semi-Supervised Model Learning

Yan Shen, Zhanghexuan Ji, Chunwei Ma, Mingchen Gao. Preprint, 2021. [Arxiv] [BibTex]