Research map / Video / Video segmentation and tracking
Lumosaic: Hyperspectral Video via Active Illumination and Coded-Exposure Pixels
Dhruv Verma, Andrew Qiu, Roberto Rangel, Ayandev Barman, Hao Yang, Chenjia Hu, Fengqi Zhang, Roman Genov, David B. Lindell, Kiriakos N. Kutulakos, Alex Mariakakis
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 |
|---|---|---|---|
| use Lumosaic sensing model instead of optical filtering, QDO or MST++ (ablation) | reconstruction backbones · Simulated HSI corpus | spectral fidelity (SAM, SSIM) | Improved |
| use HAN instead of MCAN/SRNet backbone (ablation) | Lumosaic · Simulated HSI corpus | PSNR: HAN 44.0 dB noise-free; MCAN/SRNet slightly worse but 52 ms / 27 ms per frame | Improved |
| use SPECAT backbone instead of HAN under high noise (ablation) | Lumosaic · Simulated HSI corpus (high noise) | PSNR | Hurt |
| add flow-based temporal alignment of LED sub-images (ablation) | Lumosaic · Real dynamic scenes | visual artifacts (qualitative) | Improved |
Closest papers by meaning
- Active Hyperspectral Imaging Using an Event Camera (CVPR 2025 · similarity 0.88)
- High-Speed Hyperspectral Video Acquisition By Combining Nyquist and Compressive Sampling (TPAMI 2018 · similarity 0.86)
- SASSI — Super-Pixelated Adaptive Spatio-Spectral Imaging (TPAMI 2021 · similarity 0.86)
- Exploring Spatiotemporal Feature Propagation for Video-Level Compressive Spectral Reconstruction: Dataset, Model and Benchmark (CVPR 2026 · similarity 0.86)
- l-Net: Reconstruct Hyperspectral Images From a Snapshot Measurement (ICCV 2019 · similarity 0.86)
- Dense Dispersed Structured Light for Hyperspectral 3D Imaging of Dynamic Scenes (CVPR 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.
