Research map / Image generation / Image restoration
Learning to Enhance Aperture Phasor Field for Non-Line-of-Sight Imaging
In Cho, Hyunbo Shim, Seon Joo Kim
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 denoising criterion to SSN (ablation) | SSN · Synthetic NLOS (Conf-16) | PSNR: 23.27 -> 29.55 | Improved |
| naive denoising without phasor scheme (ablation) | time-domain enhancement network · Real-world NLOS (Bike, Resolution) | visual quality | Hurt |
| add phasor-based scheme on top of denoising (ablation) | time-domain enhancement network · Synthetic NLOS (Conf-16) | PSNR: 30.69 -> 32.02 | Improved |
| add phasor-based scheme on top of denoising (ablation) | time-domain enhancement network · Synthetic NLOS (Non-16) | PSNR: 36.36 -> 37.45 | Improved |
| use band-limited phasor frequencies around the central frequency instead of low-pass or high-pass (ablation) | LEAP · Synthetic NLOS (Conf-16) | PSNR: 31.23 (low-pass) / 31.54 (high-pass) -> 32.02 | Improved |
| add attention branches (as in SSN) | signal recovery network · Synthetic NLOS | PSNR and runtime: 72.9 ms vs 20.7 ms | No clear effect |
| add signal recovery network and volume refinement (USM) vs plain LFE | LFE · Synthetic NLOS (Conf-16) | PSNR: 23.05 -> 29.99 | Improved |
| reduce exposure time per pixel | denoising variants · Real-world Bike | visual quality | Mixed |
Closest papers by meaning
- Generalizable Non-Line-of-Sight Imaging with Learnable Physical Priors (ICCV 2025 · similarity 0.88)
- Non-Line-of-Sight Imaging With Signal Superresolution Network (CVPR 2023 · similarity 0.86)
- Deep Spatial-Angular Regularization for Light Field Imaging, Denoising, and Super-Resolution (TPAMI 2021 · similarity 0.85)
- Deep Non-line-of-sight Imaging from Under-scanning Measurements (NeurIPS 2023 · similarity 0.85)
- Virtual Scanning: Unsupervised Non-line-of-sight Imaging from Irregularly Undersampled Transients (NeurIPS 2024 · similarity 0.84)
- Dual-branch Graph Feature Learning for NLOS Imaging (AAAI 2025 · similarity 0.83)
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.
