Research map / Video / Video segmentation and tracking
Exploring Spatiotemporal Feature Propagation for Video-Level Compressive Spectral Reconstruction: Dataset, Model and Benchmark
Lijing Cai, Zhan Shi, Chenglong Huang, Jinyao Wu, Qiping Li, Zikang Huo, Linsen Chen, Chongde Zi, Xun Cao
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 CDPA to the baseline (ablation) | PG-SVRT baseline · Break-down ablation | PSNR (dB): 39.97 -> 41.30 | Improved |
| add MGDP after CDPA (ablation) | PG-SVRT · Break-down ablation | PSNR (dB): 41.30 -> 41.41 | Improved |
| add MDFFN after CDPA and MGDP (ablation) | PG-SVRT · Break-down ablation | PSNR (dB): 41.41 -> 41.52 | Improved |
| add CDPA to the baseline (ablation) | PG-SVRT baseline · Break-down ablation | GFLOPs: 30.11 -> 21.11 | Improved |
| add CDPA to the baseline (ablation) | PG-SVRT baseline · Break-down ablation | STRRED: 43.90 -> 25.44 | Improved |
| add value-sharing propagation to spatial-then-temporal attention (ablation) | PG-SVRT · Spatiotemporal strategy | PSNR (dB): 41.08 -> 41.52 | Improved |
| add value-sharing propagation to temporal-then-spatial attention (ablation) | PG-SVRT · Spatiotemporal strategy | PSNR (dB): 41.04 -> 41.47 | Improved |
| use spatial-then-temporal with propagation instead of parallel processing (ablation) | PG-SVRT · Spatiotemporal strategy | PSNR (dB): 41.35 -> 41.52 | Improved |
| use DD-CASSI instead of SD-CASSI optical system | PG-SVRT · SCI system comparison | PSNR (dB): 37.78 -> 41.52 | Improved |
| use DD-CASSI instead of NDSSI | PG-SVRT · SCI system comparison | PSNR (dB): 37.84 -> 41.52 | Improved |
| use DD-CASSI instead of PMVIS | PG-SVRT · SCI system comparison | PSNR (dB): 28.45 -> 41.52 | Improved |
| use PG-SVRT instead of DPU* (temporal-concatenation variant) | DPU* · DynaSpec test | PSNR (dB): 41.36 -> 41.82 | Improved |
Closest papers by meaning
- l-Net: Reconstruct Hyperspectral Images From a Snapshot Measurement (ICCV 2019 · similarity 0.90)
- SPECAT: SPatial-spEctral Cumulative-Attention Transformer for High-Resolution Hyperspectral Image Reconstruction (CVPR 2024 · similarity 0.89)
- Improving Spectral Snapshot Reconstruction with Spectral-Spatial Rectification (CVPR 2024 · similarity 0.87)
- FRN: Fractal-Based Recursive Spectral Reconstruction Network (NeurIPS 2025 · similarity 0.87)
- DiffSCI: Zero-Shot Snapshot Compressive Imaging via Iterative Spectral Diffusion Model (CVPR 2024 · similarity 0.87)
- Hyperspectral Image Reconstruction via Combinatorial Embedding of Cross-Channel Spatio-Spectral Clues (AAAI 2024 · similarity 0.87)
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.
