atlas

Research map / Image generation / Autoregressive and latent generation

Learning-Based Sampling for Natural Image Matting

CVPR 2019 · Image generation › Autoregressive and latent generation · approach: Classical generative image synthesis · 136 citations (OpenAlex)

Check your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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
add background color predictions (ablation)Alpha matting network · Adobe datasetSAD: 47.13 -> 39.91Helped
add background color predictions (ablation)Alpha matting network · Adobe datasetMSE: 0.0146 -> 0.0115Helped
add foreground color predictions in addition to background predictions (ablation)AlphaGAN matting network · Adobe datasetMSE: 0.0115 -> 0.0099Helped
add foreground and background color predictions (ablation)Deep Matting architecture · Adobe datasetSAD: 53.56 -> 42.01Helped
add foreground and background color predictions (ablation)Deep Matting architecture · Adobe datasetMSE: 0.0187 -> 0.0114Helped
add foreground-set augmentation (ablation)AlphaGAN matting network · Adobe datasetSAD: 42.70 -> 40.35Helped
add foreground-set augmentation (ablation)AlphaGAN matting network · Adobe datasetMSE: 0.0126 -> 0.0099Helped
add alpha-gradient loss (ablation)AlphaGAN matting network · Adobe datasetMSE: 0.0107 -> 0.0099Helped
add alpha-gradient loss (ablation)AlphaGAN matting network · Adobe datasetSAD: 40.26 -> 40.35Hurt

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