Unsupervised Video Segmentation on Object-centric Video Modeling Datasets
1.2Time per frame (ms)Video-Saur
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Video-SaurHardware=NVIDIA RTX A6000 GPU, Slot number=7, Temporal context window=5 seconds, Video clip duration=30 seconds2025.01 | 1.2 | 411.703 | |
| Slot-BERTHardware=NVIDIA RTX A6000 GPU, Slot number=7, Temporal context window=5 seconds, Video clip duration=30 seconds2025.01 | 1.7 | 453.45 | |
| SAViHardware=NVIDIA RTX A6000 GPU, Slot number=7, Temporal context window=5 seconds, Video clip duration=30 seconds2025.01 | 3.3 | 430.677 | |
| Slot-DiffusionHardware=NVIDIA RTX A6000 GPU, Slot number=7, Temporal context window=5 seconds, Video clip duration=30 seconds2025.01 | 5.7 | 334.196 | |
| STEVEHardware=NVIDIA RTX A6000 GPU, Slot number=7, Temporal context window=5 seconds, Video clip duration=30 seconds2025.01 | 8.2 | 540.701 |