Research map / Image generation / Text-to-image diffusion
Real-World Image Variation by Aligning Diffusion Inversion Chain
Yuechen ZHANG, Jinbo Xing, Eric Lo, Jiaya Jia
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 |
|---|---|---|---|
| align the generation chain to the real-image inversion chain using RIVAL | Stable Diffusion V1.5 image variation · Image variation | Palette Distance: 3.650 ± 1.30 -> 1.674 ± 0.63 | Improved |
| add cross-image attention injection and attention fusion (ablation) | RIVAL · Image variation | Image Alignment: 0.760 -> 0.846 | Improved |
| add cross-image attention injection and attention fusion (ablation) | RIVAL · Image variation | Palette Distance: 2.576 -> 1.810 | Improved |
| add attention fusion (ablation) | RIVAL with attention injection · Image variation | Image Alignment: 0.818 -> 0.846 | Improved |
| add attention fusion (ablation) | RIVAL with attention injection · Image variation | Palette Distance: 1.902 -> 1.810 | Improved |
| add latent initialization alignment (ablation) | RIVAL · Image variation | Image Alignment: 0.653 -> 0.846 | Improved |
| add latent initialization alignment (ablation) | RIVAL · Image variation | Palette Distance: 3.741 -> 1.810 | Improved |
| add noise alignment (ablation) | RIVAL · Image variation | Palette Distance: 2.335 -> 1.810 | Improved |
| add noise alignment (ablation) | RIVAL · Image variation | Image Alignment: 0.839 -> 0.846 | Improved |
| remove cross-image self-attention injection (ablation) | RIVAL · Image variation | semantic and exemplar correspondence | Hurt |
| remove latent alignment and initialize from standard Gaussian noise (ablation) | RIVAL · Image variation | semantic and tone consistency | Hurt |
| use excessively long or short early alignment stages (ablation) | RIVAL · Image variation | visual quality | Hurt |
Closest papers by meaning
- Difference Inversion: Interpolate and Isolate the Difference with Token Consistency for Image Analogy Generation (CVPR 2025 · similarity 0.90)
- Noise Diffusion for Enhancing Semantic Faithfulness in Text-to-Image Synthesis (CVPR 2025 · similarity 0.90)
- Inversion-Free Image Editing with Language-Guided Diffusion Models (CVPR 2024 · similarity 0.89)
- Color Alignment in Diffusion (CVPR 2025 · similarity 0.89)
- Smooth Diffusion: Crafting Smooth Latent Spaces in Diffusion Models (CVPR 2024 · similarity 0.89)
- Inference-Time Text-to-Video Alignment with Diffusion Latent Beam Search (NeurIPS 2025 · similarity 0.89)
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.
