AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

TextDiffuser: Diffusion Models as Text Painters

Jingye Chen, Yupan Huang, Tengchao Lv, Lei Cui, Qifeng Chen, Furu Wei

NeurIPS 2023 · Image generation › Text-to-image diffusion · approach: Compositional text-to-image guidance · 15 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
Add explicit layout and character segmentation guidance to the diffusion modelTextDiffuser · MARIO-EvalOCR F-measure: 0.5865 -> 0.7824Improved
Add width embedding (ablation)Layout Transformer with 1 layer · MARIO-10M testIoU: 0.268 -> 0.289Improved
Add width embedding (ablation)Layout Transformer with 2 layers · MARIO-10M testIoU: 0.269 -> 0.298Improved
Add width embedding (ablation)Layout Transformer with 4 layers · MARIO-10M testIoU: 0.294 -> 0.297Improved
Use 2 Transformer layers rather than 1 or 4, with width embedding (ablation)Layout Transformer · MARIO-10M testIoU: 0.289, 0.298, 0.297Improved
Add character-aware loss with weight 0.01 rather than weight 0 (ablation)TextDiffuser · DrawBenchTextAccuracy: 0.396 -> 0.494Improved
Set character-aware loss weight to 0.001 rather than 0 (ablation)TextDiffuser · DrawBenchTextAccuracy: 0.396 -> 0.486Improved
Set character-aware loss weight to 0.1 rather than 0 (ablation)TextDiffuser · DrawBenchTextAccuracy: 0.396 -> 0.420No clear effect
Set character-aware loss weight to 1 rather than 0 (ablation)TextDiffuser · DrawBenchTextAccuracy: 0.396 -> 0.400No clear effect
Train whole-image and part-image generation branches together rather than train only one branch (ablation)TextDiffuser · DrawBenchText and text inpaintingAccuracy / Det-F / Spot-F: 0.494 / 0.380 / 0.218 -> 0.552 / 0.881 / 0.715Mixed
Use a 0.25 whole-image training ratio rather than 0 (ablation)TextDiffuser · DrawBenchText whole-image generationAccuracy: 0.344 -> 0.562Improved
Use a 0.75 whole-image training ratio rather than 0.5 (ablation)TextDiffuser · Text inpaintingDet-F: 0.881 -> 0.921Improved

Later papers that compared with this method

11 later accepted papers report TextDiffuser as a baseline in their results tables; 9 of them report a better number on most of the results they share with it.

3 more not listed.

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