Get Started
Home
Topics
Search
Library
Research questionHow can spectral compressive imaging models adapt to unseen optical configurations without heavy computation?Spectral compressive imaging reconstructs spectral information from compressed measurements. Deep-unfolding models can be computationally costly and may perform poorly when the camera’s optical configuration differs from training.
AI
Computer Vision
Image & Video Processing
Inference Optimization
Machine Learning
Latest papersRecent research connected to this question, newest first.SlowFast-SCI: Slow-Fast Deep Unfolding Learning for Spectral Compressive ImagingThe source concerns deep-unfolding spectral compressive imaging for cameras and spectral setups unseen during training. It reports self-supervised test-time adaptation with lightweight modules, along with reductions in parameters and FLOPs and improvements on out-of-distribution data; evidence is limited to the reported experiments and stated broader modularity.research paper · Sep 2, 2026
Related questions
How can hyperspectral image compression preserve spatial and spectral fidelity under tight storage budgets?How can diffusion priors reconstruct images from linear measurements while balancing fidelity and diversity?How does structured model compression change a model’s ability to adapt at test time under distribution shift?Can reducing the complexity of a linear generative prior improve expected reconstruction error in noiseless Gaussian compressed sensing?