aiqbench analysis
StereoEngine addresses the critical data bottleneck in robotic manipulation by synthesizing high-quality stereo pairs from monocular inputs or simulation, enabling better training for Vision-Language-Action models without relying on scarce real-world stereo datasets. This approach leverages video diffusion and rendering techniques to generate consistent depth supervision, directly advancing the scalability of bimanual robot policies through synthetic data augmentation.
StereoEngine: Scaling Stereo Bimanual Robot Data without Stereo Cameras
arXiv:2610.12734v1 Announce Type: cross Abstract: Stereo vision is increasingly used for robotic manipulation, with a growing body of work incorporating it into Vision-Language-Action (VLA) models. However, the scarcity of large-scale, real-world stereo datasets remains a fundamental bottleneck for scaling such policies. Recent generative monocular-to-stereo methods target settings such as VR content and autonomous driving. No prior work produces realistic synthetic stereo data for robotic manipulation, where stereo supervision has come predominantly from simulation. We propose StereoEngine, which introduces two complementary methods to synthesize stereo data for robotic manipulation. In the first, we warp a monocular (left) video into the right view using estimated depth and inpaint the disoccluded regions with a video diffusion model, lifting monocular demonstrations into stereo pairs without training on real stereo. In the second, we render a stereo pair in simulation and jointly generate both photorealistic views, producing consistent left-right data without any real video as input. StereoEngine outperforms existing baselines in generated stereo quality, and using its synthetic data for policy mid-training or task-specific fine-tuning substantially improves downstream bimanual manipulation. Project website: https://stereoengine.github.io
Abstract:Stereo vision is increasingly used for robotic manipulation, with a growing body of work incorporating it into Vision-Language-Action (VLA) models. However, the scarcity of large-scale, real-world stereo datasets remains a fundamental bottleneck for scaling such policies. Recent generative monocular-to-stereo methods target settings such as VR content and autonomous driving. No prior work produces realistic synthetic stereo data for robotic manipulation, where stereo supervision has come predominantly from simulation. We propose StereoEngine, which introduces two complementary methods to synthesize stereo data for robotic manipulation. In the first, we warp a monocular (left) video into the right view using estimated depth and inpaint the disoccluded regions with a video diffusion model, lifting monocular demonstrations into stereo pairs without training on real stereo. In the second, we render a stereo pair in simulation and jointly generate both photorealistic views, producing consistent left-right data without any real video as input. StereoEngine outperforms existing baselines in generated stereo quality, and using its synthetic data for policy mid-training or task-specific fine-tuning substantially improves downstream bimanual manipulation. Project website: this https URL
Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
arXiv:2610.12734 [cs.RO]
(or arXiv:2610.12734v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2610.12734 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 277 words
Related coverage
[AINews] not much happened today
Latent Space · Sep 10, 2026
Motus2: A Self-Evolving General World Model for Dexterous Manipulation
arXiv · Sep 2, 2026
NVIDIA GTC Taipei at COMPUTEX: Live Updates on What’s Next in AI
NVIDIA · May 21, 2026
HiFi-UMI: Learning Deployable Manipulation Policies from High-Fidelity UMI Data Alone
arXiv · Jul 29, 2026