AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

MatteFormer: Transformer-Based Image Matting via Prior-Tokens

GyuTae Park, SungJoon Son, JaeYoung Yoo, SeHo Kim, Nojun Kwak

CVPR 2022 · Architectures and efficiency › Attention and transformers · approach: Efficient image transformers · 87 citations (OpenAlex)

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use MatteFormer instead of SIMComposition-1k matting · Composition-1k testSAD: 28.0 -> 23.8Improved
Use MatteFormer instead of MG Matting-trimap,res50*Composition-1k matting · Composition-1k testMSE ($10^{-3}$): 5.4 -> 4.0Improved
Use MatteFormer instead of SIMComposition-1k matting · Composition-1k testGrad: 10.8 -> 8.7Improved
Use MatteFormer instead of MG Matting-trimap,res50*Composition-1k matting · Composition-1k testConn: 24.3 -> 18.9Improved
Use MatteFormer instead of MG Matting-trimap*Distinctions-646 matting · Distinctions-646 testSAD: 23.9 -> 21.9Improved
Use MatteFormer instead of MG Matting-trimap*Distinctions-646 matting · Distinctions-646 testMSE ($10^{-3}$): 7.4 -> 6.6Improved
Use MatteFormer instead of MG Matting-trimap*Distinctions-646 matting · Distinctions-646 testGrad: 14.0 -> 11.2Improved
Use MatteFormer instead of MG Matting-trimap*Distinctions-646 matting · Distinctions-646 testConn: 22.4 -> 20.5Improved
Add a GAP prior-token (ablation)Swin Transformer Tiny matting baseline · Composition-1kSAD: 26.43 -> 25.30Improved
Replace the GAP token with an unknown-region prior-token (ablation)Swin Transformer Tiny matting baseline · Composition-1kSAD: 25.30 -> 24.70Improved
Add foreground and background prior-tokens alongside the unknown-region prior-token (ablation)Swin Transformer Tiny matting baseline · Composition-1kSAD: 24.70 -> 24.19Improved
Add prior-memory to the three trimap-region prior-tokens (ablation)Swin Transformer Tiny matting baseline · Composition-1kSAD: 24.19 -> 23.80Improved

Later papers that compared with this method

1 later accepted paper reports MatteFormer as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy