Measure Twice, Cut Once: Grasping Video Structures and Event Semantics with LLMs for Video Temporal Localization
Open paper
Benchmark
Model
Rank
Results
moment-retrieval-on-charades-sta
MeCo (QWen2VL 7B) (extra data)
#15
R@1 IoU=0.5: 68.5
R@1 IoU=0.7: 41.6
R@1 IoU=0.3: 82.3
Rank counts only results with a code link.