Greg Shakhnarovich

Toyota Technological Institute at Chicago

Papers

Total Citations

H-Index

About

Greg Shakhnarovich is a prominent computer vision and machine learning researcher whose work spans self-supervised learning, geometric scene understanding, and practical robotics perception. His research addresses some of the most pressing challenges in autonomous systems, particularly the development of scalable, cost-effective alternatives to expensive sensing hardware like LiDAR. Among his notable contributions, Shakhnarovich has advanced the frontier of self-supervised depth estimation, most prominently through his work on full surround monodepth from multiple cameras — a significant leap beyond traditional single-camera or stereo approaches that accumulates 51 citations since 2022. This work enables richer environmental perception critical for autonomous driving applications. Complementing this, his research on self-supervised camera self-calibration from video (27 citations) tackles the tedious and costly process of camera calibration, proposing elegant solutions that learn directly from raw video streams without specialized data collection. Shakhnarovich's contributions are particularly impactful for the robotics and autonomous driving communities, where reducing dependence on expensive hardware while maintaining geometric accuracy is a central challenge. His body of work reflects a consistent vision: making robust, geometry-aware perception accessible, scalable, and practical for real-world deployment.