AI Paper Atlas

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

Implicit Style-Content Separation using B-LoRA

Yarden Frenkel, Yael Vinker, Ariel Shamir, Danny Cohen-Or

ECCV 2024 · Image generation › Text-to-image diffusion · approach: Personalized and region-controlled generation · 104 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
train LoRA only on blocks 4 and 5 (B-LoRA) instead of all blocks to separate style and content (ablation)B-LoRA on SDXL · image stylizationstorage reduction: 70% less storageImproved
optimize blocks {4,5} instead of {2,5} (ablation)B-LoRA on SDXL · single-image reconstructionqualitative reconstruction and content captureImproved
use the prompt 'A [v]' instead of 'A [v] <class-name>' during B-LoRA training (ablation)B-LoRA on SDXL · style transfer between objectsqualitative content preservation and style transferImproved
inject 'A [c]' into content block W4 and 'A [c] in [s] style' into other blocks at inference (ablation)B-LoRA on SDXL · complex scenesqualitative generation qualityImproved
scale style adapter W5 by alpha 0.4-0.5 (ablation)B-LoRA on SDXL · identity preservationqualitative color preservationImproved
use B-LoRA instead of DB-LoRA + ControlNetstyle transfer · Style transfer (multiple content images)DINO style similarity: 0.863 -> 0.881Improved
use B-LoRA instead of ZipLoRA on content preservationstyle transfer · Content preservation (multiple content images)DINO content similarity: 0.841 -> 0.790Hurt
use B-LoRA instead of SWAG (non-SDXL) for style transferstyle transfer · Style transfer (single image)DINO style similarity: 0.883 -> 0.881No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 2 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