Research map / Image generation / Image restoration
Similarity-Consistent Likelihood Diffusion enables Hidden Person Detection from Wall Reflections
Zhiwen Zheng, Hao Zhou, Huiyu Qi, Zhao Huang, Guangyuan Zhang, Shaowei Jiang, Wenwen Tang, Bin Yang, Jin Liu, Xiaoshuai Zhang, Xingru Huang
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 DeLi-Inversion pseudo-likelihood guided diffusion (full SLD-Net) to DeLi-only output (ablation) | SLD-Net · Reflect-Corridor | FID: 217.30 -> 73.54 | Improved |
| add DeLi likelihood to diffusion-only sampling (ablation) | SLD-Net · Reflect-Corridor | PSNR: 6.02 -> 15.58 | Improved |
| add diffusion prior to DeLi-only output (ablation) | SLD-Net · Reflect-Room | PSNR: 11.14 -> 12.49 | Improved |
| add DeLi likelihood to diffusion-only sampling (ablation) | SLD-Net · Reflect-Room | FID: 182.98 -> 26.89 | Improved |
| replace CFG (scale 1.0) with precision-weighted Bayesian guidance (ablation) | SLD-Net · Reflect-Corridor | PSNR: 13.42 -> 15.58 | Improved |
| replace CFG (scale 1.0) with precision-weighted Bayesian guidance (ablation) | SLD-Net · Reflect-Corridor | FID: 53.17 -> 73.54 | Hurt |
| increase CFG scale from 1.0 to 7.0 (ablation) | DDIM+CFG · Reflect-Corridor | FID: 53.17 -> 114.39 | Hurt |
| add DeLi pixel-wise precision field Lambda_0 to mean-only fusion (ablation) | SLD-Net · Reflect-Corridor | PSNR: 13.48 -> 15.58 | Improved |
| add DeLi pixel-wise precision field Lambda_0 to mean-only fusion (ablation) | SLD-Net · Reflect-Corridor | FID: 53.31 -> 73.54 | Hurt |
| reduce DDIM steps from 100 to 25 (ablation) | SLD-Net · Reflect-Corridor and Reflect-Room | PSNR/SSIM | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Restormer — Restormer: Efficient Transformer for High-Resolution Image Restoration (CVPR 2022) · better on 2 of 2 shared results
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
- Soft Shadow Diffusion (SSD): Physics-inspired Learning for 3D Computational Periscopy (ECCV 2024 · similarity 0.88)
- Using Unknown Occluders to Recover Hidden Scenes (CVPR 2019 · similarity 0.86)
- Steady-State Non-Line-Of-Sight Imaging (CVPR 2019 · similarity 0.85)
- Efficient Non-Line-of-Sight Imaging from Transient Sinograms (ECCV 2020 · similarity 0.85)
- Virtual Scanning: Unsupervised Non-line-of-sight Imaging from Irregularly Undersampled Transients (NeurIPS 2024 · similarity 0.85)
- Adaptive 3D Reconstruction via Diffusion Priors and Forward Curvature-Matching Likelihood Updates (NeurIPS 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.
