Novel Object Detection On Lvis V1 0 Val
Metrics
All mAP
Known mAP
Novel mAP
Results
Performance results of various models on this benchmark
Model Name | All mAP | Known mAP | Novel mAP | Paper Title | Repository |
|---|---|---|---|---|---|
| UNO Fini et al. (2021)++ | 2.18 | 21.09 | 0.61 | A Unified Objective for Novel Class Discovery | |
| Cooperative Foundational Models | 19.33 | 42.08 | 17.42 | Enhancing Novel Object Detection via Cooperative Foundational Models | |
| Weng et al. Weng et al. (2021)* | 1.62 | 17.85 | 0.27 | Unsupervised Discovery of the Long-Tail in Instance Segmentation Using Hierarchical Self-Supervision | - |
| RNCDL | 6.92 | 25.00 | 5.42 | Learning to Discover and Detect Objects | |
| ORCA Cao et al. (2022) | 2.03 | 20.57 | 0.49 | Open-World Semi-Supervised Learning |
0 of 5 row(s) selected.