AI Paper Atlas

Research map / Image generation / Editing and style transfer

The Image Local Autoregressive Transformer

Chenjie Cao, Yuxin Hong, Xiang Li, Chengrong Wang, Chengming Xu, Yanwei Fu, Xiangyang Xue

NeurIPS 2021 · Image generation › Editing and style transfer · approach: Image-to-image translation · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

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 iLAT with local discrete representations and the LA attention mask instead of Taminglocally guided pose synthesis · Penn Action (PA)PSNR: 21.43 -> 22.94Improved
Use iLAT with local discrete representations and the LA attention mask instead of Taminglocally guided pose synthesis · Penn Action (PA)FID: 33.53 -> 27.36Improved
Use iLAT instead of Taminglocally guided image synthesis · Synthetic DeepFashion (SDF)SSIM: 0.539 -> 0.599Improved
Use iLAT instead of Taminglocally guided image synthesis · Synthetic DeepFashion (SDF)FID: 72.77 -> 70.58Improved
Generate only masked regions autoregressivelyiLAT · PAinference time (sec/image): 8.551 -> 3.426Improved
Generate only masked regions autoregressivelyiLAT · FFHQinference time (sec/image): 8.183 -> 1.180Improved
Add two-stream convolutions to prevent encoder information leakage (ablation)iLAT · PAFID: 31.83 -> 27.36Improved
Add two-stream convolutions to prevent encoder information leakage (ablation)iLAT · PAPSNR: 21.68 -> 22.94Improved
Add two-stream convolutions to prevent encoder information leakage (ablation)iLAT · PASSIM: 0.748 -> 0.800Improved
Add two-stream convolutions to prevent encoder information leakage (ablation)iLAT · PAMAE: 0.056 -> 0.046Improved
Use the LA attention mask with Taming rather than its vanilla attention maskTaming · Face editingaverage human evaluation score: 0.33 -> 3.33Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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