AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

PhotoFramer: Multi-modal Image Composition Instruction

Zhiyuan You, Ke Wang, He Zhang, Xin Cai, Jinjin Gu, Tianfan Xue, Chao Dong, Zhoutong Zhang

CVPR 2026 · Vision-language and multimodal › Human and embodied modeling · approach: Vision-language pretraining · 7 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
finetune Bagel on the composition triplet datasetBagel · Shift task benchmark, GPT-5 win rate vs originalwin rate (%): 27.61 -> 80.37Improved
finetune Bagel on the composition dataset (text-image consistency)Bagel · Text-image guidance consistency, averageconsistency score: 83.10 -> 92.02Improved
use PhotoFramer instead of open-source editing models (Qwen-Image-Edit)image editing · Zoom-in task benchmark, GPT-5 win rate vs ground truthwin rate (%): 39.65 -> 67.24Improved
use PhotoFramer vs gpt-image-1 on view-changeimage generation · View-change task, GPT-5 win rate vs originalwin rate (%): 84.61 vs 82.07No clear effect
allow free viewpoint changes (view-change task) for open-source editorsQwen-Image-Edit · View-change vs shift task, GPT-5 win rate vs originalwin rate (%): 46.01 -> 70.65Improved
feed PhotoFramer text guidance into Qwen-Image-Edit (ablation)Qwen-Image-Edit · text-guided editingfidelity / compositionHurt
feed text guidance into gpt-image-1 (ablation)gpt-image-1 · text-guided editingcompositionImproved
train with text guidance instead of image pairs only (ablation)Bagel · qualitative (Fig 11)compositionImproved
train Bagel on text + image instead of finetuning Kontext on images only (ablation)finetuned editor · qualitative (Fig 12)compositionImproved
use auto prompt (shift and/or zoom-in) instead of single task prompt (ablation)PhotoFramer · qualitative (Fig 9)compositionImproved
train composition assessment 7B model with GRPO vs Qwen2.5-VL-32Bassessment model · GAIC composition assessmentSRCC: 0.195 -> 0.795Improved

Later papers that compared with this method

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

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