Igor Vasiljevic

Toyota Technological Institute at Chicago, Toyota Research Institute

Papers

Total Citations

H-Index

About

Igor Vasiljevic is a computer vision and robotics researcher whose work sits at the intersection of self-supervised learning, 3D scene understanding, and autonomous systems. He has made significant contributions to the problem of depth estimation and sensor calibration without relying on expensive labeled data or hardware like LiDAR. His most cited work, "Full Surround Monodepth From Multiple Cameras" (2022, 51 citations), extended self-supervised monocular depth estimation beyond single cameras to full 360-degree perception—a critical advance for real-world autonomous driving applications. Complementing this, his research on self-supervised camera self-calibration (2022, 27 citations) tackled the tedious and brittle process of manual camera calibration, proposing methods that learn geometric parameters directly from video streams. His subsequent work on robust extrinsic self-calibration further strengthened multi-camera systems for deployment in diverse, unpredictable environments. More recently, Vasiljevic has expanded into neural radiance fields with DeLiRa, addressing limitations of volumetric rendering under constrained viewpoints, and into language-grounded 3D reasoning with Transcrib3D, reflecting a growing interest in enabling robots to interpret natural human instructions within complex spatial environments.