Quanta AI Labs

Google DeepMind · SfM Benchmark

ORBIT++

An expanded version of the ORBIT benchmark from CVPR 2026 for evaluating structure-from-motion and camera-pose estimation in the wild, built by using 360-degree panoramic video as a source to synthesise challenging perspective clips while keeping ground-truth camera trajectories recoverable. It evaluates MegaSaM, COLMAP and VGGT-Long across six challenge categories — high speed, low texture, low light, crowds, egocentric parallel moving objects and fluids — on ATE and RPE-R, finding low-texture and high-speed scenes hardest overall. Code and dataset released.

Open the source ↗

Similar 3D & Spatial releases

  • Mundo: Customized World Model

    Mundo project team · Customised world model · Week 45, 2026

  • Idol

    Speridlabs (Alejandro Cañada, Chema Garabito) · Image-to-3D splats · Week 45, 2026

  • JST-2

    Viggle AI · World Model · Week 44, 2026

  • PURE

    Peking University, HKUST, Pengcheng Laboratory · 3D Reconstruction · Week 44, 2026

  • Motion2MixamoRig

    MaxLiu22 · Video Motion Retargeting · Week 43, 2026

  • InfiniSplat v2

    Zhejiang University, Shenzhen University, Hello TransTech and Causa Robotics · Feed-Forward Gaussian Splatting · Week 43, 2026

More from Google DeepMind

Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.