Weakly Supervised Object Detection On Comic2K
Metriken
MAP
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Vergleichstabelle
| Modellname | MAP |
|---|---|
| h2fa-r-cnn-holistic-and-hierarchical-feature | 53.0 |
| domain-adaptive-self-supervised-pre-training | 67.41 |
| cross-domain-weakly-supervised-object | 42.2 |
| multiple-instance-learning-on-deep-features | 27 |
| informative-and-consistent-correspondence | 37.1 |
| h2fa-r-cnn-holistic-and-hierarchical-feature | 46.4 |
| cross-domain-weakly-supervised-object | 37.2 |
| text-image-alignment-for-diffusion-based | 57.4 |