Aleksandr Khorin

Moscow Institute of Physics and Technology

Papers

Total Citations

H-Index

About

Aleksandr Khorin is a rising researcher in the field of autonomous driving perception, with a primary focus on 3D LiDAR point cloud segmentation and domain adaptation. His work addresses a critical bottleneck in deploying self-driving vehicles: the inability of segmentation models trained on synthetic or source-domain data to generalize to real-world, target-domain environments. Khorin’s most notable contribution, "DAPS3D: Domain Adaptive Projective Segmentation of 3D LiDAR Point Clouds" (2023), introduces a novel projective framework that bridges the domain gap by aligning feature representations across different sensor configurations and environmental conditions. This approach has already garnered 13 citations, signaling its importance to the community. By tackling the challenge of segmenting key scene elements—such as roads, buildings, pedestrians, and vehicles—from sparse, irregular point clouds, Khorin’s work directly enhances the reliability of environmental ranging for autonomous systems. His research is particularly impactful for students and engineers seeking robust, real-world solutions for LiDAR-based perception, offering a pathway toward more adaptable and safer autonomous vehicles.