Natural Language Queries on Ego4D-NLQ v2 (test)
27.6Recall@1 (IoU=0.3)OSGNet
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| OSGNetFeature=E+I2025.05 | 27.6 | 19.78 | 43.46 | 32.77 | |
| GroundVQAFeature=E+I, Published=CVPR 2024, NaQ pretraining=true, reproduced=true2025.05 | 26.67 | 17.63 | 39.94 | 27.7 | |
| EgoVideoFeature=EgoVideo, Published=CVPRW 2024, NaQ pretraining=true2025.05 | 25.07 | 17.31 | 40.88 | 29.67 | |
| GroundNLQFeature=E+I, Published=CVPRW 2023, NaQ pretraining=true2025.05 | 24.5 | 17.31 | 40.46 | 29.17 | |
| ASLFeature=E+I, Published=CVPRW 2023, NaQ pretraining=true2025.05 | 24.13 | 15.46 | 34.37 | 23.18 | |
| EgoEnvFeature=E+I+C+EgoEnv, Published=NeurIPS 2024, NaQ pretraining=true2025.05 | 23.28 | 14.36 | 27.25 | 17.58 | |
| NaQFeature=E+I+C, Published=CVPR 2023, NaQ pretraining=true, reproduced=true2025.05 | 21.7 | 13.64 | 25.12 | 16.33 |