AI Paper Atlas

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

Make-a-Scene: Scene-Based Text-to-Image Generation with Human Priors

Oran Gafni, Adam Polyak, Oron Ashual, Shelly Sheynin, Devi Parikh, Yaniv Taigman

ECCV 2022 · Image generation › Text-to-image diffusion · approach: Personalized and region-controlled generation · 436 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 scene tokens (ablation)base transformer · MS-COCO valFID: 18.01 -> 19.16Hurt
add scene tokens (human preference) (ablation)base transformer · MS-COCO valhuman preference: 57.3% image quality, 65.3% photorealism, 58.3% text alignmentImproved
add face-aware VQ loss (ablation)+Scene tokens · MS-COCO valFID: 19.16 -> 14.45Improved
add classifier-free guidance (ablation)+Face-aware · MS-COCO valFID: 14.45 -> 7.55Improved
add object-aware VQ at 512x512 (ablation)+CF · MS-COCO valFID: 7.55 -> 8.70Hurt
add object-aware VQ at 512x512 (human preference) (ablation)+CF · MS-COCO valhuman preference: 62.0% quality, 53.5% photorealism, 52.2% text alignmentMixed
give scene as additional input (ablation)Make-A-Scene with CF · MS-COCO valFID: 7.55 -> 4.69Improved
use Make-A-Scene instead of LAFITEtext-to-image · MS-COCO valFID: 8.12 -> 7.55Improved
use Make-A-Scene instead of GLIDE (filtered)text-to-image · MS-COCO val (filtered)FID: 12.24 -> 11.84Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 5 of 5 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