Loading...
Loading...
Available on 1 platform
Sign in to view source links and access this dataset
Reflect3R is a synthetic dataset for training models to perform 3D stereo reconstruction from a single camera view, aided by mirror reflections. The data was created by researchers from the University of Oxford and is associated with a paper for the 3DV 2026 conference. It was last updated on the Hugging Face platform in March 2026.
License is unknown; users must verify terms before use.