HyperAIHyperAI
AccueilActualitésArticles de rechercheTutorielsEnsembles de donnéesWikiSOTAModèles LLMClassement GPUÉvénements
Recherche
À propos
HyperAIHyperAI

Highlight Detection On Tvsum

Métriques

mAP

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle
mAP
Paper TitleRepository
FlashVTG88FlashVTG: Feature Layering and Adaptive Score Handling Network for Video Temporal Grounding
SG-DETR87.1Saliency-Guided DETR for Moment Retrieval and Highlight Detection
UVCOM (train from scratch)86.3Bridging the Gap: A Unified Video Comprehension Framework for Moment Retrieval and Highlight Detection
QD-DETR86.6Query-Dependent Video Representation for Moment Retrieval and Highlight Detection
QD-DETR (only Video)85.0Query-Dependent Video Representation for Moment Retrieval and Highlight Detection
UMT83.1UMT: Unified Multi-modal Transformers for Joint Video Moment Retrieval and Highlight Detection
CG-DETR86.8Correlation-Guided Query-Dependency Calibration for Video Temporal Grounding
0 of 7 row(s) selected.