EviDETR: Preserving Query-Relevant Temporal Evidence for Moment Retrieval and Highlight Detection

arXiv:2609.30724v1 Announce Type: new Abstract: Joint video moment retrieval and highlight detection requires identifying query-relevant temporal segments while estimating clip-level saliency, yet DETR-style pipelines do not explicitly preserve query-relevant evidence throughout encoding, decoding, and cross-task prediction. We propose EviDETR, an evidence-preserving framework with three components. Semantic-aware Feature Reweighting (SFR) enhances query-relevant clip representations through saliency estimation and cross-modal interaction. A Temporal Top-2 Mixture-of-Experts (TTop2MoE) decoder performs query-adaptive refinement via sparse expert routing. MR-to-HD (MR2HD) fusion transfers span-level retrieval evidence to clip-level highlight prediction through confidence-weighted multi-scale aggregation. Using CLIP+SlowFast features, EviDETR achieves 69.29 R1@0.5, 54.77 R1@0.7, and 48.41 Avg. mAP for moment retrieval on QVHighlights, together with 41.83 HD-mAP and 68.33 HIT@1. Strong results on TACoS and Charades-STA further demonstrate cross-dataset transferability.

Related

• Concentrate After Imagination: Text-Conditioned Evidence Grounding for Partially Relevant Video Retrieval
• REZE: Recognition-Based Zero-Shot Extraction for Video Temporal Grounding
• Ground, Cover, and Refine: Evidence-Centric Frame Selection for Long-Video Question Answering

Source: arXiv cs.CV | 2026-09-28