Reproduce “Multi-Band Galaxy Novel-View Synthesis with Radiative Gaussian Fields” 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
Yuzhe Wu, Jialin Zhu, Doğa Yılmaz, Daisuke Kawata, He Wang
GalaxyGS reconstructs unseen galaxy views from sparse multi-band images using small 3D blobs whose contributions render images. It separates stellar emission, dust emission, and dust attenuation into distinct Gaussian fields and fits them in stages. On Auriga data, its SED error is 0.068 ± 0.036, versus 0.140 ± 0.059 for the best baseline; this is not a single-view real-galaxy test.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.