Research map / Image generation / Autoregressive and latent generation
MISC: Multi-Condition Injection and Spatially-Adaptive Compositing for Conditional Person Image Synthesis
Shuchen Weng, Wenbo Li, Dawei Li, Hongxia Jin, Boxin Shi
Check 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 geometry-guided gates to AdaIN via GAIN (ablation) | MISC pattern injection · VIP test | FID: 18.03 -> 16.09 | Improved |
| Add geometry-guided gates to AdaIN via GAIN (ablation) | MISC pattern injection · VIP test | M-score: 12.54 -> 3.86 | Improved |
| Replace uniform SPADE pattern injection with GAIN (ablation) | MISC pattern injection · VIP test | FID: 16.72 -> 16.09 | Improved |
| Add cross-modality ranking loss LCM (ablation) | MISC color injection · VIP test | R-precision (%): 83.89 -> 93.59 | Improved |
| Use spatially-specific instead of uniform color injection (ablation) | MISC color injection · VIP test | R-precision (%): 76.71 -> 83.89 | Improved |
| Use spatially-specific color injection rather than uniform injection with auxiliary classifier (ablation) | MISC color injection · VIP test | FID: 41.59 -> 16.09 | Improved |
| Add the compositing model (ablation) | MISC · VIP test | FID: 22.09 -> 16.09 | Improved |
| Add the compositing model (ablation) | MISC · VIP test | M-score: 151.39 -> 3.86 | Improved |
| Use foreground-only alpha-blending compositing instead of joint foreground-background translation (ablation) | MISC compositing · VIP test | FID: 16.86 -> 16.09 | Improved |
| Add bounded tanh to spatially-adaptive compositing (ablation) | MISC compositing · VIP test | M-score: 33.64 -> 3.86 | Improved |
| Use spatially-adaptive rather than spatially-invariant bounded compositing (ablation) | MISC compositing · VIP test | FID: 52.23 -> 16.09 | Improved |
| Use spatially-adaptive rather than spatially-invariant unbounded compositing (ablation) | MISC compositing · VIP test | M-score: 161.07 -> 33.64 | Improved |
Later papers that compared with this method
1 later accepted paper reports MISC as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- UniCoRN: A Unified Conditional Image Repainting Network (CVPR 2022 · 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
- IMAGPose: A Unified Conditional Framework for Pose-Guided Person Generation (NeurIPS 2024 · similarity 0.88)
- Advancing Pose-Guided Image Synthesis with Progressive Conditional Diffusion Models (ICLR 2024 · similarity 0.87)
- Training-free Dense-Aligned Diffusion Guidance for Modular Conditional Image Synthesis (CVPR 2025 · similarity 0.84)
- ONE-SHOT: Compositional Human-Environment Video Synthesis via Spatial-Decoupled Motion Injection and Hybrid Context Integration (NeurIPS 2026 · similarity 0.84)
- Multi-focal Conditioned Latent Diffusion for Person Image Synthesis (CVPR 2025 · similarity 0.84)
- Anywhere: A Multi-Agent Framework for User-Guided, Reliable, and Diverse Foreground-Conditioned Image Generation (AAAI 2025 · similarity 0.84)
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.
