Benchmark RadarStereoGenBench
StereoGenBench is a synthetic multi-camera benchmark for stereo generation, rendered in Unreal Engine with a rigid six-camera array. It provides calibrated view pairs across baseline regimes, along with RGB, metric depth, intrinsics, and poses for each scene.
- Released
- 2026-05-22
- Readiness
- Inspectable
- Primary field
- General AI
Why it matters
Stereo generation and view synthesis require controlled baseline and intrinsics for evaluation, which existing resources lack. This benchmark enables measuring sensitivity to baseline regimes and target-camera consistency under matched scene content, supporting development of stereo generation models.
Motivation
Stereo image and video generation, stereo geometry estimation, and condition-controlled view synthesis require paired data in which the variables that determine binocular geometry -- camera baseline, intrinsics, scene depth, and camera motion -- are known and controllable.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.