AI Paper Atlas

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

Lego: Learning to Disentangle and Invert Personalized Concepts Beyond Object Appearance in Text-to-Image Diffusion Models

Saman Motamed, Danda Pani Paudel, Luc Van Gool

ECCV 2024 · Image generation › Text-to-image diffusion · approach: Personalized and region-controlled generation · 17 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
invert concept with Lego instead of describing it in natural languageLDM · Concept preference user study (7 non-numeric concepts)human preference: preferred over 70% of casesImproved
invert cardinality concept 4 with Lego rather than natural language promptsLDM · Cardinality concept 4images with correct count: 18 -> 63Improved
invert walking on a rope with Lego rather than natural languageLDM · Walking on a rope (VQA)images confirmed by Flan-T5 XL: 4 -> 133Improved
add Subject Separation (explicit subject embedding learned from concept-free images) (ablation)Lego / Textual Inversion · burnt and melted, closed eyes, walking on ropequalitative concept fidelityImproved
add contrastive Context Loss with synonym and antonym words (ablation)Lego · burnt and melted, closed eyes, walking on ropequalitative concept fidelityImproved
add Context Loss without Subject Separation (ReVersion-like) (ablation)Textual Inversion · subject-entangled conceptsqualitative concept fidelityNo clear effect
fine-tune diffusion model parts (Custom-Diffusion) to invert entangled conceptsCustom-Diffusion · melting, closed eyesappearance leakageHurt
increase concept images from 2 to 4 (ablation)Lego · closed eyes (VQA accuracy)accuracy: 75% -> 83%Improved
increase concept images from 4 to 8 (ablation)Lego · closed eyes (VQA accuracy)accuracy: 83% -> 83%No clear effect
use Lego-learned subject embedding in place of Textual Inversion for subject synthesisLego · subject synthesis user studypreference rate: 55% -> 45%No clear effect

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