Reproduce “You Only Encode Once: Sub-Pixel-Shifted Frames for Arbitrary-Scale Remote Sensing Super-Resolution with 2D Gaussian Splatting” 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
Canjie Liu, Liuqian Wang, Qi Ming, Wei Lu, Xiaohan Hao, Yuefei Wu, Wentao Chao, Lei Shi
SPS-GS reconstructs remote-sensing images at multiple scales by combining several shifted low-resolution observations before one deep encoding pass, then refining a shared 2D Gaussian representation against the observations. On AID at ×8, an unseen training scale, it reaches 27.52 dB PSNR versus 26.76 dB for ST-AVSR. These results are conditional on the tested synthesized-input and real-world evaluation protocols.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.