Video Summarization on MoSu
0.351Tau (τ)TripleSumm
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| TripleSummModality=V, T, A, Params=1.37M, GFLOPs=0.97G2026.03 | 0.351 | 0.472 | 74.72 | 44.42 | |
| CSTAModality=V, Params=10.56M, GFLOPs=11.37G2026.03 | 0.291 | 0.398 | 71.77 | 40.65 | |
| CFSumModality=V, T, A, Params=19.83M, GFLOPs=8.52G2026.03 | 0.277 | 0.374 | 70.97 | 38.2 | |
| UMTModality=V, T, Params=4.66M, GFLOPs=1.39G2026.03 | 0.239 | 0.334 | 68.83 | 36.73 | |
| SSPVSModality=V, T, Params=112.81M, GFLOPs=43.64G2026.03 | 0.19 | 0.271 | 66.1 | 32.65 | |
| Joint-VAModality=V, A, Params=4.21M, GFLOPs=1.63G2026.03 | 0.19 | 0.272 | 65.68 | 32.25 | |
| A2SummModality=V, T, Params=2.48M, GFLOPs=1.35G2026.03 | 0.181 | 0.257 | 66.48 | 35.7 | |
| VASNetModality=V, Params=8.13M, GFLOPs=1.99G2026.03 | 0.151 | 0.219 | 64.49 | 31.05 | |
| PGL-SUMModality=V, Params=5.31M, GFLOPs=1.21G2026.03 | 0.151 | 0.218 | 64.97 | 30.63 |