Open Vocabulary Multi-Label Classification with Dual-Modal Decoder on Aligned Visual-Textual Features

Benchmark Model Rank Results
multi-label-classification-on-ms-cocoADDS(ViT-L-336, resolution 1344)–mAP: 93.54
multi-label-classification-on-ms-cocoADDS(ViT-L-336, resolution 336)–mAP: 91.76
multi-label-classification-on-ms-cocoADDS(ViT-L-336, resolution 640)–mAP: 93.41
multi-label-zero-shot-learning-on-nus-wideADDS (ViT-B-32, resolution 224)–mAP: 36.56
multi-label-zero-shot-learning-on-nus-wideADDS (ViT-L-336, resolution 336)–mAP: 39.01