VISTA: Validation-Guided Integration of Spatial and Temporal Foundation Models with Anatomical Decoding for Rare-Pathology VCE Event Detection -- after competition results
Computer Vision and Pattern Recognition2026-05-22v1
Capsule endoscopy event detection is challenging because clinically relevant findings are sparse, visually heterogeneous, and evaluated at the event level rather than by frame accuracy. We propose VISTA, a metric-aligned multi-backbone framework for the RAREVISION task. VISTA combines EndoFM-LV for temporal context and DINOv3 ViTL/16 for frame-level visual semantics, followed by a Diverse Head Ensemble (DHE), Validation-Guided Weighted Fusion (VGWF), and Anatomy-Aware Temporal Event Decoding (ATED). The original official submission achieved hidden-test temporal [email protected] of 0.3530 and [email protected] of 0.3235. After the competition, extending local threshold refinement with a global coarse search improved performance to 0.3726 [email protected] and 0.3431 [email protected], ranking Team ACVLab second in the post-competition evaluation.
@article{arxiv.2605.22096,
title = {VISTA: Validation-Guided Integration of Spatial and Temporal Foundation Models with Anatomical Decoding for Rare-Pathology VCE Event Detection -- after competition results},
author = {Bo-Cheng Qiu and Fang-Ying Lin and Ming-Han Sun and Yu-Fan Lin and Chia-Ming Lee and Chih-Chung Hsu},
journal= {arXiv preprint arXiv:2605.22096},
year = {2026}
}