Video Highlight Detection on Mr.HiSum
86mAP (rho=50%)HiVid
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| HiVidModel Category=Large Model-based2026.02 | 86 | — | 53 | — | — | 0.66 | 0.67 | |
| MLPModel Category=Vision Saliency-based2026.02 | 81 | — | 49 | — | — | 0.59 | 0.6 | |
| SL-moduleModel Category=Vision Saliency-based2026.02 | 81 | — | 49 | — | — | 0.59 | 0.6 | |
| DETRModel Category=Vision Saliency-based2026.02 | 81 | — | 45 | — | — | 0.57 | 0.58 | |
| PGL-SUMModel Category=Vision Saliency-based2026.02 | 80 | — | 46 | — | — | 0.52 | 0.54 | |
| VASNetModel Category=Vision Saliency-based2026.02 | 80 | — | 46 | — | — | 0.55 | 0.56 | |
| VideoLLaMA3Model Category=Large Model-based2026.02 | 77 | — | 45 | — | — | 0.54 | 0.55 | |
| VILAModel Category=Large Model-based2026.02 | 73 | — | 44 | — | — | 0.52 | 0.54 | |
| TripleSumm (Full)Modality=Visual, Text, Audio2026.03 | 70.72 | — | 40.88 | — | — | — | — | |
| DAViHDModality=video and audio2026.02 | 67.27 | 59.73 | 36.55 | 0.299 | 0.213 | — | — | |
| TripleSumm (Visual)Modality=Visual2026.03 | 67.16 | — | 35.57 | — | — | — | — | |
| UMTModality=Visual, Text2026.03 | 66.81 | — | 35.65 | — | — | — | — | |
| Joint-VAModality=Visual, Text2026.03 | 65.88 | — | 35.23 | — | — | — | — | |
| UMTModality=video and audio2026.02 | 65.81 | 58.18 | 33.79 | 0.239 | 0.174 | — | — | |
| CSTAModality=Visual2026.03 | 63.38 | — | 30.42 | — | — | — | — | |
| A2SummModality=Visual, Text2026.03 | 63.2 | — | 32.34 | — | — | — | — | |
| Joint-VAModality=video and audio2026.02 | 61.82 | 54.71 | 29.09 | 0.152 | 0.104 | — | — | |
| PGL-SUMModality=Visual2026.03 | 61.6 | — | 27.45 | — | — | — | — | |
| CSTAModality=video only2026.02 | 61.12 | 54.32 | 28.35 | 0.138 | 0.095 | — | — | |
| PGL-SUMModality=video only2026.02 | 59.73 | 53.34 | 25.71 | 0.104 | 0.07 | — | — | |
| SSPVSModality=Visual, Text2026.03 | 59.48 | — | 26.35 | — | — | — | — | |
| VASNetModality=Visual2026.03 | 58.69 | — | 25.28 | — | — | — | — | |
| FlamingoModel Category=Large Model-based2026.02 | 56 | — | 33 | — | — | 0.41 | 0.41 |