Reproduce “Deep Decomposition Learning for Inverse Imaging Problems” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Dongdong Chen, Mike E. Davies
The authors introduce DDN, a physics-engaged network for reconstructing images from incomplete measurements. It learns separate corrections for measurement-visible range information and unmeasured nullspace information, then combines them with a backprojected estimate. In simulated CS-MRF with 10 training samples, DDN’s NMSE was 1.056 × 10⁻² versus 1.119 × 10⁻² for nullspace learning; this result does not establish clinical performance.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.