SG-EBM

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 ParsingonVisual Genome
Recall@50· 2021-03-03
31.74
best: 32.9 (SpeaQ (without reweighting))
Energy-Based Learning for Scene Graph Generation arXiv:2103.02221
Scene ParsingonVisual Genome
mean Recall @20· 2021-03-03
7.1
best: 12.3 (ADTrans)
Energy-Based Learning for Scene Graph Generation arXiv:2103.02221
Scene Graph GenerationonVisual Genome
Recall@50· 2021-03-03
31.74
best: 32.9 (SpeaQ (without reweighting))
Energy-Based Learning for Scene Graph Generation arXiv:2103.02221
Scene Graph GenerationonVisual Genome
mean Recall @20· 2021-03-03
7.1
best: 12.3 (ADTrans)
Energy-Based Learning for Scene Graph Generation arXiv:2103.02221

Audio2 results

2D Semantic SegmentationonVisual Genome
Recall@50· 2021-03-03
31.74
best: 32.9 (SpeaQ (without reweighting))
Energy-Based Learning for Scene Graph Generation arXiv:2103.02221
2D Semantic SegmentationonVisual Genome
mean Recall @20· 2021-03-03
7.1
best: 12.3 (ADTrans)
Energy-Based Learning for Scene Graph Generation arXiv:2103.02221