Research map / Image generation / Editing and style transfer
ACID-Style: An Adaptive Condition Injection Diffusion Model for Arbitrary Style Transfer
Ting Yang, Siyu Yang, Xiyao Liu, Songtao Wu, Gerald Schaefer, Kuanhong Xu, Hui Fang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| train injection modules with the diffusion loss instead of content/style losses (ablation) | ACID-Style · Style transfer ablation | style score (CSD): 0.4825 -> 0.0974 | Hurt |
| use adaptive content/style loss weights tied to the diffusion schedule instead of constant weights (ablation) | ACID-Style · Style transfer ablation | style score (CSD): 0.3159 -> 0.4825 | Improved |
| use adaptive loss weights instead of constant weights (ablation) | ACID-Style · Style transfer ablation | Ls (style loss): 1.1059 -> 0.8853 | Improved |
| use adaptive loss weights instead of constant weights (ablation) | ACID-Style · Style transfer ablation | Lc (content loss): 1.5537 -> 1.5017 | Improved |
| use adaptive loss weights instead of constant weights (ablation) | ACID-Style · Style transfer ablation | content score (CLIP): 0.7989 -> 0.7753 | Hurt |
| increase lambda from 0.2 to 0.8 (ablation) | ACID-Style · Style transfer lambda sweep | style score (CSD): 0.2160 -> 0.5774 | Improved |
| increase lambda from 0.2 to 0.8 (ablation) | ACID-Style · Style transfer lambda sweep | Lc (content loss): 0.8150 -> 1.8638 | Hurt |
| use lightweight adapters on frozen SD instead of inversion-based or per-style-trained diffusion methods | ACID-Style · Inference time | time per image (s): 6.07 -> 1.79 (vs CSGO); 1976.3 (InST) | Improved |
| ACID-Style versus Attn. Distill. and StyleSSP on style score and ArtFID | ACID-Style · Style transfer (800 images) | ArtFID: 19.0274 (Attn. Distill.) vs 21.1256 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 3 on most shared results.
- StyTR2 — StyTr2: Image Style Transfer With Transformers (CVPR 2022) · better on 3 of 3 shared results
- StyleSSP — StyleSSP: Sampling StartPoint Enhancement for Training-free Diffusion-based Method for Style Transfer (CVPR 2025) · better on 0 of 1 shared result
- CSGO — CSGO: Content-Style Composition in Text-to-Image Generation (NeurIPS 2025) · better on 1 of 1 shared result
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
- Style Injection in Diffusion: A Training-free Approach for Adapting Large-scale Diffusion Models for Style Transfer (CVPR 2024 · similarity 0.91)
- StyleSSP: Sampling StartPoint Enhancement for Training-free Diffusion-based Method for Style Transfer (CVPR 2025 · similarity 0.88)
- Z*: Zero-shot Style Transfer via Attention Reweighting (CVPR 2024 · similarity 0.88)
- SigStyle: Signature Style Transfer via Personalized Text-to-Image Models (AAAI 2025 · similarity 0.88)
- ArtBank: Artistic Style Transfer with Pre-trained Diffusion Model and Implicit Style Prompt Bank (AAAI 2024 · similarity 0.87)
- Conditional Balance: Improving Multi-Conditioning Trade-Offs in Image Generation (CVPR 2025 · similarity 0.87)
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.
