18 papers
PatchGen: Learning Soft Intra-Image Predictive Subsets for Visual Generalization
Zhaorui Tan, Weimiao Yu, Xi Yang
Visual classifiers are expected to generalize under data shifts, target shifts, and their combinations, yet most existing methods focus on domain invariance while failing to addres…
Beyond Instance-Level Self-Supervision in 3D Multi-Modal Medical Imaging
Tan Pan, Shuhao Mei, Yixuan Sun +8
Self-supervised pre-training methods in medical imaging typically treat each individual as an isolated instance, learning representations through augmentation-based objectives or m…
Minimal Semantic Sufficiency Meets Unsupervised Domain Generalization
Tan Pan, Kaiyu Guo, Dongli Xu +8
The generalization ability of deep learning has been extensively studied in supervised settings, yet it remains less explored in unsupervised scenarios. Recently, the Unsupervised…
Saving for the future: Enhancing generalization via partial logic regularization
Zhaorui Tan, Yijie Hu, Xi Yang +3
Generalization remains a significant challenge in visual classification tasks, particularly in handling unknown classes in real-world applications. Existing research focuses on the…
Exploiting Layer Normalization Fine-tuning in Visual Transformer Foundation Models for Classification
Zhaorui Tan, Tan Pan, Kaizhu Huang +8
LayerNorm is pivotal in Vision Transformers (ViTs), yet its fine-tuning dynamics under data scarcity and domain shifts remain underexplored. This paper shows that shifts in LayerNo…
Structure-aware Semantic Discrepancy and Consistency for 3D Medical Image Self-supervised Learning
Tan Pan, Zhaorui Tan, Kaiyu Guo +6
3D medical image self-supervised learning (mSSL) holds great promise for medical analysis. Effectively supporting broader applications requires considering anatomical structure var…
Disentangling Tabular Data Towards Better One-Class Anomaly Detection
Jianan Ye, Zhaorui Tan, Yijie Hu +3
Tabular anomaly detection under the one-class classification setting poses a significant challenge, as it involves accurately conceptualizing "normal" derived exclusively from a si…
Covariance-based Space Regularization for Few-shot Class Incremental Learning
Yijie Hu, Guanyu Yang, Zhaorui Tan +3
Few-shot Class Incremental Learning (FSCIL) presents a challenging yet realistic scenario, which requires the model to continually learn new classes with limited labeled data (i.e.…
Towards a Universal 3D Medical Multi-modality Generalization via Learning Personalized Invariant Representation
Zhaorui Tan, Xi Yang, Tan Pan +8
Variations in medical imaging modalities and individual anatomical differences pose challenges to cross-modality generalization in multi-modal tasks. Existing methods often concent…
Interpret Your Decision: Logical Reasoning Regularization for Generalization in Visual Classification
Zhaorui Tan, Xi Yang, Qiufeng Wang +2
Vision models excel in image classification but struggle to generalize to unseen data, such as classifying images from unseen domains or discovering novel categories. In this paper…
CathAction: A Benchmark for Endovascular Intervention Understanding
Baoru Huang, Tuan Vo, Chayun Kongtongvattana +35
Real-time visual feedback from catheterization analysis is crucial for enhancing surgical safety and efficiency during endovascular interventions. However, existing datasets are of…
MedMAP: Promoting Incomplete Multi-modal Brain Tumor Segmentation with Alignment
Tianyi Liu, Zhaorui Tan, Muyin Chen +3
Brain tumor segmentation is often based on multiple magnetic resonance imaging (MRI). However, in clinical practice, certain modalities of MRI may be missing, which presents a more…
Revisiting Mutual Information Maximization for Generalized Category Discovery
Zhaorui Tan, Chengrui Zhang, Xi Yang +2
Generalized category discovery presents a challenge in a realistic scenario, which requires the model's generalization ability to recognize unlabeled samples from known and unknown…
SCMix: Stochastic Compound Mixing for Open Compound Domain Adaptation in Semantic Segmentation
Kai Yao, Zhaorui Tan, Zixian Su +3
Open compound domain adaptation (OCDA) aims to transfer knowledge from a labeled source domain to a mix of unlabeled homogeneous compound target domains while generalizing to open…
Rethinking Information Loss in Medical Image Segmentation with Various-sized Targets
Tianyi Liu, Zhaorui Tan, Kaizhu Huang +1
Medical image segmentation presents the challenge of segmenting various-size targets, demanding the model to effectively capture both local and global information. Despite recent e…
Rethinking Multi-domain Generalization with A General Learning Objective
Zhaorui Tan, Xi Yang, Kaizhu Huang
Multi-domain generalization (mDG) is universally aimed to minimize the discrepancy between training and testing distributions to enhance marginal-to-label distribution mapping. How…
Semantic-aware Data Augmentation for Text-to-image Synthesis
Zhaorui Tan, Xi Yang, Kaizhu Huang
Data augmentation has been recently leveraged as an effective regularizer in various vision-language deep neural networks. However, in text-to-image synthesis (T2Isyn), current aug…
SSD: Towards Better Text-Image Consistency Metric in Text-to-Image Generation
Zhaorui Tan, Xi Yang, Zihan Ye +4
Generating consistent and high-quality images from given texts is essential for visual-language understanding. Although impressive results have been achieved in generating high-qua…