Reproduce “Image Guided Depth Upsampling using Anisotropic Total Generalized Variation” 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
David Ferstl, Christian Reinbacher, Rene Ranftl, Matthias Rüther, Horst Bischof
The authors present a depth-upscaling method combining sparse depth measurements with a higher-resolution intensity image. A convex optimization uses second-order Total Generalized Variation and an image-derived anisotropic tensor to guide smoothing. On Middlebury’s ×8 Art test with added Gaussian noise, pixel-disparity RMSE is 2.06 versus 2.78 for Park et al.; this result does not guarantee performance on arbitrary real scenes.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.