AI Paper Atlas

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

PartCraft: Crafting Creative Objects by Parts

Kam Woh Ng, Xiatian Zhu, Yi-Zhe Song, Tao Xiang

ECCV 2024 · Image generation › Text-to-image diffusion · approach: Layout-to-image control · 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
add the token bottleneck MLP encoder (ablation)PartCraft · CUB-200-2011FID: 16.36 -> 12.86Improved
replace MSE attention loss with entropy-based normalized attention loss (ablation)PartCraft · CUB-200-2011 part compositionEMR and CoSimImproved
add bottleneck encoder to speed up learning (ablation)PartCraft · convergence analysislearning speedImproved
increase lambda_attn from 0.01 to 0.1 (ablation)PartCraft · CUB-200-2011EMR: 0.460 -> 0.339Hurt
decrease lambda_attn from 0.01 to 0.00001 (ablation)PartCraft · CUB-200-2011FID: 12.86 -> 11.64 (EMR 0.460 -> 0.397)Mixed
fine-tune the diffusion model (DreamBooth) instead of textual inversion onlyTextual Inversion · CUB-200-2011EMR: 0.305 -> 0.355Improved
unconstrained fine-tuning of cross-attention K/V (Custom Diffusion)Textual Inversion · CUB-200-2011FID: 10.10 -> 37.61Hurt
use PartCraft instead of Break-a-Scenepart reconstruction · CUB-200-2011EMR: 0.390 -> 0.460Improved

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