Research map / Architectures and efficiency / RNNs and normalization
Binarized Spectral Compressive Imaging
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| replace vanilla 1-bit convolution with BiSR-Conv (ablation) | binarized base model · KAIST simulation | PSNR (dB): 23.90 -> 27.80 | Helped |
| use the four binarized modules (BiDS, BiUS, BiFD, BiFU) instead of normal binarized modules (ablation) | BiSRNet · KAIST simulation | PSNR (dB): 27.80 -> 29.76 | Helped |
| add spectral redistribution (SR) to BiSR-Conv (ablation) | Baseline-2 · KAIST simulation | PSNR (dB): 27.68 -> 28.97 | Helped |
| add Tanh(alpha x) Sign approximation (ablation) | Baseline-2 · KAIST simulation | PSNR (dB): 27.68 -> 28.74 | Helped |
| combine SR and Tanh(alpha x) (ablation) | Baseline-2 · KAIST simulation | PSNR (dB): 27.68 -> 29.76 | Helped |
| replace Clip(x) with Tanh(alpha x) for backward Sign approximation (ablation) | BiSRNet · KAIST simulation | PSNR (dB): 28.97 -> 29.76 | Helped |
| replace Quad(x) with Tanh(alpha x) for backward Sign approximation (ablation) | BiSRNet · KAIST simulation | PSNR (dB): 29.02 -> 29.76 | Helped |
| binarize only the bottleneck vs full-precision model (ablation) | base model · KAIST simulation | PSNR (dB): 34.11 -> 33.80 | Hurt |
| binarize only the decoder vs full-precision model (ablation) | base model · KAIST simulation | PSNR (dB): 34.11 -> 33.03 | Hurt |
| BiSRNet vs best prior BNN BTM | 1-bit HSI reconstruction · KAIST simulation | avg PSNR (dB): 27.21 -> 29.76 | Helped |
Closest papers by meaning
- In2SET: Intra-Inter Similarity Exploiting Transformer for Dual-Camera Compressive Hyperspectral Imaging (CVPR 2024 · similarity 0.87)
- HIR-Diff: Unsupervised Hyperspectral Image Restoration Via Improved Diffusion Models (CVPR 2024 · similarity 0.87)
- SPECAT: SPatial-spEctral Cumulative-Attention Transformer for High-Resolution Hyperspectral Image Reconstruction (CVPR 2024 · similarity 0.87)
- Untrained Neural Nets for Snapshot Compressive Imaging: Theory and Algorithms (NeurIPS 2024 · similarity 0.87)
- Detail Matters: Mamba-Inspired Joint Unfolding Network for Snapshot Spectral Compressive Imaging (AAAI 2025 · similarity 0.86)
- Dual-Window Multiscale Transformer for Hyperspectral Snapshot Compressive Imaging (AAAI 2024 · similarity 0.86)
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.