Research map / Image generation / Editing and style transfer
StereoEdit: A Diffusion-Based Framework for Stereo-Consistent Image Editing
Baolin Liu, Zongyuan Yang, Yingde Song, yongping xiong
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 |
|---|---|---|---|
| add GGAR to base SD 1.5 (ablation) | SD 1.5 · Stereo editing set (120 pairs) ablation | Warp-Err: 8.78 -> 4.41 | Improved |
| add GGAR to base SD 1.5 (ablation) | SD 1.5 · Stereo editing set (120 pairs) ablation | CLIP-F: 94.97 -> 97.23 | Improved |
| add GALA to base SD 1.5 (ablation) | SD 1.5 · Stereo editing set (120 pairs) ablation | Warp-Err: 8.78 -> 5.15 | Improved |
| add GALA on top of GGAR (ablation) | StereoEdit · Stereo editing set (120 pairs) ablation | Warp-Err: 4.41 -> 3.92 | Improved |
| add GALA on top of GGAR (ablation) | StereoEdit · Stereo editing set (120 pairs) ablation | CLIP-T: 33.19 -> 33.82 | Improved |
| add reliability log modulation to GGAR (ablation) | StereoEdit · Stereo editing set (120 pairs) ablation | Warp-Err: 4.55 -> 4.41 | Improved |
| replace video-style attention alignment (WAttn) with GGAR alongside GALA (ablation) | StereoEdit · Stereo editing set (120 pairs) ablation | Warp-Err: 5.04 -> 3.92 | Improved |
| replace WAttn with GGAR alongside GALA (ablation) | StereoEdit · Stereo editing set (120 pairs) ablation | CLIP-F: 96.31 -> 97.36 | Improved |
| use StereoEdit instead of best prior method on Warp-Err | StereoEdit · Stereo editing set (120 pairs) | Warp-Err: 5.53 -> 3.92 | Improved |
| use StereoEdit instead of VideoGrain on CLIP-T | StereoEdit · Stereo editing set (120 pairs) | CLIP-T: 33.77 -> 33.82 | No clear effect |
| use StereoEdit instead of VidToMe on CLIP-F | StereoEdit · Stereo editing set (120 pairs) | CLIP-F: 96.38 -> 97.36 | Improved |
| use StereoEdit instead of VideoGrain on human Overall | StereoEdit · Human evaluation (50 pairs, 1-10) | Overall: 7.18 -> 8.09 | Improved |
Closest papers by meaning
- Edit2Perceive: Image Editing Diffusion Models Are Strong Dense Perceivers (CVPR 2026 · similarity 0.90)
- SyncNoise: Geometrically Consistent Noise Prediction for Instruction-based 3D Editing (AAAI 2025 · similarity 0.88)
- ProEdit: Simple Progression is All You Need for High-Quality 3D Scene Editing (NeurIPS 2024 · similarity 0.88)
- EditSplat: Multi-View Fusion and Attention-Guided Optimization for View-Consistent 3D Scene Editing with 3D Gaussian Splatting (CVPR 2025 · similarity 0.87)
- ConsistDreamer: 3D-Consistent 2D Diffusion for High-Fidelity Scene Editing (CVPR 2024 · similarity 0.87)
- Diffusion Handles Enabling 3D Edits for Diffusion Models by Lifting Activations to 3D (CVPR 2024 · 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.
