HiLo(R50)

Reported on 6 benchmarks across 3 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision4 results

Scene ParsingonPSG Dataset
R@20· 2023-03-28
34.1
best: 40.6 (HiLo(SwinL))
HiLo: Exploiting High Low Frequency Relations for Unbiased Panoptic Scene Graph Generation arXiv:2303.15994
Scene ParsingonPSG Dataset
mR@20· 2023-03-28
23.7
best: 34.7 (VLPrompt (R50))
HiLo: Exploiting High Low Frequency Relations for Unbiased Panoptic Scene Graph Generation arXiv:2303.15994
Scene Graph GenerationonPSG Dataset
R@20· 2023-03-28
34.1
best: 40.6 (HiLo(SwinL))
HiLo: Exploiting High Low Frequency Relations for Unbiased Panoptic Scene Graph Generation arXiv:2303.15994
Scene Graph GenerationonPSG Dataset
mR@20· 2023-03-28
23.7
best: 34.7 (VLPrompt (R50))
HiLo: Exploiting High Low Frequency Relations for Unbiased Panoptic Scene Graph Generation arXiv:2303.15994

Audio2 results

2D Semantic SegmentationonPSG Dataset
R@20· 2023-03-28
34.1
best: 40.6 (HiLo(SwinL))
HiLo: Exploiting High Low Frequency Relations for Unbiased Panoptic Scene Graph Generation arXiv:2303.15994
2D Semantic SegmentationonPSG Dataset
mR@20· 2023-03-28
23.7
best: 34.7 (VLPrompt (R50))
HiLo: Exploiting High Low Frequency Relations for Unbiased Panoptic Scene Graph Generation arXiv:2303.15994