Flexible Techniques for Differentiable Rendering with 3D Gaussians

August 28, 2023 · Entered Twilight · 🏛 arXiv.org

"No code URL or promise found in abstract"
"Derived repo from GitHub Pages (backfill)"

Evidence collected by the PWNC Scanner

Repo contents: .gitignore, LICENSE, README.md, data, fm_render.py, generate_inputs.ipynb, get_co3d.sh, images_to_video.sh, pose_estimation.ipynb, requirements.txt, run_co3d_sp-zpfm.ipynb, run_co3d_sp.ipynb, util.py, util_load.py, util_render.py, utils_opt.py, zpfm_render.py

Authors Leonid Keselman, Martial Hebert arXiv ID 2308.14737 Category cs.CV: Computer Vision Cross-listed cs.AI, cs.GR Citations 32 Venue arXiv.org Repository https://github.com/leonidk/fmb-plus ⭐ 187 Last Checked 9 days ago

Abstract

Fast, reliable shape reconstruction is an essential ingredient in many computer vision applications. Neural Radiance Fields demonstrated that photorealistic novel view synthesis is within reach, but was gated by performance requirements for fast reconstruction of real scenes and objects. Several recent approaches have built on alternative shape representations, in particular, 3D Gaussians. We develop extensions to these renderers, such as integrating differentiable optical flow, exporting watertight meshes and rendering per-ray normals. Additionally, we show how two of the recent methods are interoperable with each other. These reconstructions are quick, robust, and easily performed on GPU or CPU. For code and visual examples, see https://leonidk.github.io/fmb-plus