GitHub

DAPS: Improving Diffusion Inverse Problem Solving with Decoupled Noise Annealing (CVPR 2025 Oral)

We propose a new method called Decoupled Annealing Posterior Sampling (DAPS) that relies on a novel noise annealing process to solve posterior sampling with diffusion prior. Specifically, we decouple ...
IEEE

Matrix-Inversion-Free Sparse Bayesian Learning for High-Resolution Radar Imagery via Improved Variational Inference

Abstract: Sparse Bayesian learning (SBL) is an advanced statistical framework that dominantly enhances the sparse features of targets of interest in radar imagery. A widely adopted strategy for ...
www.cs.utexas.edu

Multistep Inverse Is Not All You Need

Multistep Inverse Is Not All You Need. Alexander Levine, Peter Stone, and Amy Zhang. @Article{alexander_levine_RLC_2024, author = {Alexander Levine and Peter Stone and Amy Zhang}, title = {Multistep ...
GitHub

Frequency differencing method that generates low-frequency data for kickstarting full-waveform inversion

Ultrasound tomography (UST) is a medical imaging system that uses the transmission of ultrasound through tissue to create images of the speed of sound. Currently, the main algorithm used to provide ...
IEEE

A Low-computation Matrix Inverse Method based on RDR Recursion and Iterative Extended Inverse

Abstract: In space-time joint adaptive anti-jamming implementation, such as MVDR, LCMV, etc., the inversion of the covariance matrix is a critical step and the computation load is large. In this paper ...