Self-Enhanced Image Clustering with Cross-Modal Semantic Consistency

Open paper
Benchmark Model Rank Results
image-clustering-on-cifar-10SEIC (CLIP ViT-B/16)–Accuracy: 0.978NMI: 0.945ARI: 0.952
image-clustering-on-cifar-10SEIC (CLIP ViT-B/32)–Accuracy: 0.976NMI: 0.938ARI: 0.947
image-clustering-on-imagenet-10SEIC (CLIP ViT-B/16)–NMI: 0.992Accuracy: 0.997ARI: 0.994
image-clustering-on-imagenet-10SEIC (CLIP ViT-B/32)–NMI: 0.988Accuracy: 0.996ARI: 0.991
image-clustering-on-stl-10SEIC (CLIP ViT-B/16)–Accuracy: 0.993NMI: 0.980ARI: 0.985
image-clustering-on-stl-10SEIC (CLIP ViT-B/32)–Accuracy: 0.987NMI: 0.963ARI: 0.970

Rank counts only results with a code link.