AI Paper Atlas

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

LOTS of Fashion! Multi-Conditioning for Image Generation via Sketch-Text Pairing

Federico Girella, Davide Talon, Ziyue Liu, Zanxi Ruan, Yiming Wang, Marco Cristani

ICCV 2025 · Image generation › Text-to-image diffusion · approach: Compositional text-to-image guidance · 4 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
use frozen DINOv2 sketch encoder instead of end-to-end trained encoder (ablation)LOTS · Sketchy testSSIM: .615 -> .678Improved
use DINOv2 instead of CLIP sketch encoder (ablation)LOTS · Sketchy testSSIM: .623 -> .678Improved
use DINOv2 instead of BLIP2 sketch encoder (ablation)LOTS · Sketchy testSSIM: .630 -> .678Improved
add Pair-Former pooling instead of no pooling (ablation)LOTS · Sketchy testSSIM: .623 -> .678Improved
add Pair-Former pooling instead of no pooling (ablation)LOTS · Sketchy testVQAScore: .724 -> .749Mixed
defer merging to diffusion instead of mean pooling pairs (ablation)LOTS · Sketchy testVQAScore: .676 -> .749Improved
defer merging to diffusion instead of mean pooling pairs (ablation)LOTS · Sketchy testSSIM: .659 -> .678Improved
use Pair-Former self-attention instead of cross-attention (ablation)LOTS · Sketchy testVQAScore: .681 -> .749Improved
use 32 pooling tokens instead of 64 (ablation)LOTS · Sketchy testVQAScore: .669 -> .749Improved
use localized sketch-text pairs (LOTS) instead of fine-tuned ControlNetControlNet · Sketchy testGlobalCLIP: .645 -> .679Improved
use LOTS instead of fine-tuned IP-AdapterIP-Adapter · Sketchy testFID: 0.69 -> 0.79Hurt
use LOTS instead of Multi-T2I-AdapterMulti-T2I-Adapter (zero-shot) · Sketchy testSSIM: .723 -> .678Hurt

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