Lightweight unrolled deep network with self-supervised adaptation for high dynamic range recovery from modulo measurements.
Self-supervised image reconstruction and denoising framework extending recorrupted-to-recorrupted beyond Gaussian noise.
Optimization-based unlimited sampling recovery formulated via regularization and ADMM.
Deep learning framework for joint non-linear representation and compressive spectral reconstruction.
Time-reversed scene reconstruction from thermal traces using Visual-Language Models and constrained diffusion.
Preprint: a multimodal benchmark and reconstruction framework for inferring recent human-environment interactions from residual physical traces.
Preprint: scene-driven adaptive ordering of Hadamard basis patterns for single-pixel spectral imaging.
Time-reversed scene reconstruction from thermal traces using Visual-Language Models and constrained diffusion to recover past frames up to 120 seconds earlier.
Preprint: noise distribution agnostic self-supervised image denoising via learned recorruption.