MetaFormer is a general architecture abstracted from Transformers by not specifying the token mixer.
Source: MetaFormer Is Actually What You Need for VisionPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Image Classification | 8 | 16.67% |
Semantic Segmentation | 8 | 16.67% |
Object Detection | 6 | 12.50% |
Image Segmentation | 3 | 6.25% |
Medical Image Segmentation | 3 | 6.25% |
Specificity | 2 | 4.17% |
Decoder | 2 | 4.17% |
Autonomous Driving | 1 | 2.08% |
Object | 1 | 2.08% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |