Universal Domain Adaptation On Office 31
Metrics
H-score
Source-Free
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | H-score | Source-Free |
|---|---|---|
| geometric-anchor-correspondence-mining-with | 87.6 | no |
| ovanet-one-vs-all-network-for-universal | 86.5 | no |
| lead-learning-decomposition-for-source-free | 87.8 | yes |
| upcycling-models-under-domain-and-category | 87.8 | yes |
| universal-domain-adaptation-via-compressive | 95.95 | no |
| boosting-novel-category-discovery-over-1 | 91.8 | no |
| domain-consensus-clustering-for-universal | 80.16 | no |
| learning-to-detect-open-classes-for-universal | 73.14 | no |
| universal-domain-adaptation | 63.46 | no |
| unified-optimal-transport-framework-for | 91.13 | no |
| umad-universal-model-adaptation-under-domain | 87.0 | yes |
| mlnet-mutual-learning-network-with | 92.8 | no |