Multi Label Classification On Ms Coco

评估指标

mAP

评测结果

各个模型在此基准测试上的表现结果

Paper TitleRepository
ADDS(ViT-L-336, resolution 1344)93.54Open Vocabulary Multi-Label Classification with Dual-Modal Decoder on Aligned Visual-Textual Features-
ADDS(ViT-L-336, resolution 640)93.41Open Vocabulary Multi-Label Classification with Dual-Modal Decoder on Aligned Visual-Textual Features-
ADDS(ViT-L-336, resolution 336)91.76Open Vocabulary Multi-Label Classification with Dual-Modal Decoder on Aligned Visual-Textual Features-
ML-Decoder(TResNet-XL, resolution 640)91.4ML-Decoder: Scalable and Versatile Classification Head
Q2L-CvT(ImageNet-21K pretraining, resolution 384)91.3Query2Label: A Simple Transformer Way to Multi-Label Classification
MLD-TResNet-L-AAM[640x640]91.30Combining Metric Learning and Attention Heads For Accurate and Efficient Multilabel Image Classification
ML-Decoder(TResNet-L, resolution 640)91.1ML-Decoder: Scalable and Versatile Classification Head
Q2L-SwinL(ImageNet-21K pretraining, resolution 384)90.5Query2Label: A Simple Transformer Way to Multi-Label Classification
IDA-SwinL90.3Causality Compensated Attention for Contextual Biased Visual Recognition-
CCD-SwinL90.3Contextual Debiasing for Visual Recognition With Causal Mechanisms-
Q2L-TResL(ImageNet-21K pretraining, resolution 640)90.3Query2Label: A Simple Transformer Way to Multi-Label Classification
MlTr-XL(ImageNet-21K pretraining, resolution 384)90.0MlTr: Multi-label Classification with Transformer
TResNet-L-V2, (ImageNet-21K-P pretraining, resolution 640)89.8ImageNet-21K Pretraining for the Masses
MlTr-L(ImageNet-21K pretraining, resolution 384)88.5MlTr: Multi-label Classification with Transformer
TResNet-XL (resolution 640)88.4Asymmetric Loss For Multi-Label Classification
TResNet-L-V2, (ImageNet-21K-P pretraining, resolution 448)88.4ImageNet-21K Pretraining for the Masses
GKGNet(resolution 576)87.7GKGNet: Group K-Nearest Neighbor based Graph Convolutional Network for Multi-Label Image Recognition
M3TR(ImageNet-21K-P pretraining, resolution 448)87.5M3TR: Multi-modal Multi-label Recognition with Transformer-
GKGNet(resolution 448)86.7GKGNet: Group K-Nearest Neighbor based Graph Convolutional Network for Multi-Label Image Recognition
TResNet-L (resolution 448)86.6Asymmetric Loss For Multi-Label Classification
0 of 34 row(s) selected.
Multi Label Classification On Ms Coco | SOTA | HyperAI超神经