Object Detection on LVIS v1 (eval)
27.5APTFA-ours-aug
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| TFA-ours-augPretrain=All, Base Data=True, Novel Data=True, Re-training=True2022.03 | 27.5 | 19.3 | 25.6 | 33 | |
| Joint-trainPretrain=All, Base Data=True, Novel Data=True, Re-training=True2022.03 | 27.2 | 18 | 26.4 | 33.6 | |
| TFA*-augPretrain=All, Base Data=K shots, Novel Data=True, Re-training=True2022.03 | 27 | 19.7 | 27.6 | 29.6 | |
| TFA-ours-augPretrain=Aug, Base Data=True, Novel Data=True, Re-training=True2022.03 | 25.1 | 17.8 | 22.6 | 30.9 | |
| SylphPretrain=All, Continuous=True2022.03 | 24.6 | 16.5 | 23.7 | 29.1 | |
| TFA*-augPretrain=Aug, Base Data=K shots, Novel Data=True, Re-training=True2022.03 | 24.4 | 16.1 | 24.9 | 27.6 | |
| Joint-trainPretrain=Aug, Base Data=True, Novel Data=True, Re-training=True2022.03 | 24.3 | 13.3 | 22.7 | 30.9 | |
| TFA*Pretrain=Default, Base Data=K shots, Novel Data=True, Re-training=True2022.03 | 21.1 | 9.1 | 21.6 | 25.9 | |
| TFA-oursPretrain=Default, Base Data=True, Novel Data=True, Re-training=True2022.03 | 21 | 11.9 | 17.7 | 28.6 | |
| Joint-trainPretrain=Default, Base Data=True, Novel Data=True, Re-training=True2022.03 | 20.7 | 8.7 | 18.2 | 28.8 | |
| SylphPretrain=Aug, Continuous=True2022.03 | 20.7 | 13.9 | 19 | 25.5 | |
| ONCE*Pretrain=All, Continuous=True2022.03 | 19.4 | 12.3 | 18.8 | 23.3 | |
| SylphPretrain=Default, Continuous=True2022.03 | 18.5 | 10 | 16.5 | 24.3 | |
| ONCE*Pretrain=Default, Continuous=True2022.03 | 12.9 | 6.3 | 11.2 | 17.7 | |
| ONCE*Pretrain=Aug, Continuous=True2022.03 | 8.5 | 6.1 | 7.8 | 10.2 |