works on

From the 1 of 16 linked papers with an AI index.

activity
20242026
collaborators
Showing cs.CVShow all

14 papers · 1 filter

cs.CV2026

CRISP: Constrained Refinement via Iterative Squeezing Process for Robust Medical Image Segmentation under Domain Shift

Yizhou Fang, Pujin Cheng, Yixiang Liu +2

The paper introduces CRISP, a model‑agnostic framework that refines medical image segmentation without any test‑time adaptation or target data by exploiting stable probability rank…

cs.CV2026

CRISP: Rank-Guided Iterative Squeezing for Robust Medical Image Segmentation under Domain Shift

Yizhou Fang, Pujin Cheng, Yixiang Liu +2

Distribution shift in medical imaging remains a central bottleneck for the clinical translation of medical AI. Failure to address it can lead to severe performance degradation in u…

cs.CV2026

4DEquine: Disentangling Motion and Appearance for 4D Equine Reconstruction from Monocular Video

Jin Lyu, Liang An, Pujin Cheng +2

4D reconstruction of equine family (e.g. horses) from monocular video is important for animal welfare. Previous mainstream 4D animal reconstruction methods require joint optimizati…

cs.CV2025

UniFS: Unified Multi-Contrast MRI Reconstruction via Frequency-Spatial Fusion

Jialin Li, Yiwei Ren, Kai Pan +4

Recently, Multi-Contrast MR Reconstruction (MCMR) has emerged as a hot research topic that leverages high-quality auxiliary modalities to reconstruct undersampled target modalities…

cs.CV2025

AniMer+: Unified Pose and Shape Estimation Across Mammalia and Aves via Family-Aware Transformer

Liang An, Jin Lyu, Li Lin +3

In the era of foundation models, achieving a unified understanding of different dynamic objects through a single network has the potential to empower stronger spatial intelligence.…

cs.CV2025

Boosting Memory Efficiency in Transfer Learning for High-Resolution Medical Image Classification

Yijin Huang, Pujin Cheng, Roger Tam +1

The success of large-scale pre-trained models has established fine-tuning as a standard method for achieving significant improvements in downstream tasks. However, fine-tuning the…