Research map / Architectures and efficiency / Quantization
Quantization-Aware Deep Optics for Diffractive Snapshot Hyperspectral Imaging
Lingen Li, Lizhi Wang, Weitao Song, Lei Zhang, Zhiwei Xiong, Hua Huang
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 |
|---|---|---|---|
| model DOE fabrication quantization during joint DOE and reconstruction optimization | Diffractive snapshot hyperspectral imaging on ICVL · ICVL test, 4 levels | PSNR: 31.68 -> 33.89 | Improved |
| adapt the physical height assigned to each quantization level (ablation) | QDO on ICVL · ICVL test, 4 levels | PSNR: 33.89 -> 35.32 | Improved |
| replace conventional full-precision DOE optimization with quantization-aware optimization | Diffractive snapshot hyperspectral imaging · ICVL test, 4 levels | SSIM: 0.939 -> 0.962 | Improved |
| add adaptive quantization to QDO (ablation) | QDO on ICVL · ICVL test, 4 levels | RMSE: 0.0250 -> 0.0205 | Improved |
| use QDO+A rather than CASSI encoding | Snapshot hyperspectral imaging · ICVL test | PSNR: 30.66 -> 35.32 | Improved |
| use QDO+A rather than Fresnel encoding | Snapshot hyperspectral imaging · ICVL test | PSNR: 27.41 -> 35.32 | Improved |
| reduce the DOE quantization levels from 32 to 2 (ablation) | DO on ICVL · ICVL test | PSNR: 36.75 -> 28.56 | Hurt |
| reduce the DOE quantization levels from 32 to 2 (ablation) | QDO on ICVL · ICVL test | PSNR: 36.80 -> 33.42 | Hurt |
| reduce the DOE quantization levels from 32 to 2 (ablation) | QDO+A on ICVL · ICVL test | PSNR: 36.82 -> 33.97 | Hurt |
Closest papers by meaning
- Non-Serial Quantization-Aware Deep Optics for Snapshot Hyperspectral Imaging (TPAMI 2024 · similarity 0.91)
- From Local to Global: A Progressive Reconstruction Network for Diffractive Snapshot Spectral Imaging (ECCV 2026 · similarity 0.87)
- Single-Shot Hyperspectral-Depth Imaging With Learned Diffractive Optics (ICCV 2021 · similarity 0.87)
- Aperture Diffraction for Compact Snapshot Spectral Imaging (ICCV 2023 · similarity 0.87)
- OpticalNet: An Optical Imaging Dataset and Benchmark Beyond the Diffraction Limit (CVPR 2025 · similarity 0.84)
- Dense Dispersed Structured Light for Hyperspectral 3D Imaging of Dynamic Scenes (CVPR 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.
