Language-guided Video Object Segmentation on A2D (test)
57.8Precision @0.5RefVOS
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| RefVOSTraining dataset=RefCOCO, Fine-tuning=A2D2020.10 | 57.8 | 9.3 | 67.2 | 49.7 | |
| Wang et al.Pre-training=Kinetics2020.10 | 55.7 | 2 | 60.1 | 49 | |
| Gavrilyuk et al.Pre-training=Kinetics2020.10 | 50 | 0.4 | 55.1 | 42.6 | |
| RefVOSTraining dataset=A2D2020.10 | 49.5 | 6.4 | 59.9 | 43 | |
| RefVOSTraining dataset=RefCOCO2020.10 | 27.9 | 3.4 | 41.4 | 25.6 |