Scholar
Ajinkya Khoche
Google Scholar ID: BjRps2cAAAAJ
Industrial PhD | Scania | KTH Royal Institute of Technology, Stockholm
Citations & Impact
Citations
27
H-index
3
i10-index
1
Publications
9
Co-authors
6
list available
Publications
Resume (English only)
Academic Achievements
- Paper 'SSF: Sparse Long-Range Scene Flow for Autonomous Driving' accepted to ICRA 2025; code released.
- Paper 'HiMo: High-Speed Objects Motion Compensation in Point Clouds' accepted to IEEE Transactions on Robotics (T-RO).
- Paper 'Towards long-range 3D object detection for autonomous vehicles' accepted to IEEE IV 2024; presented in Jeju, South Korea.
- Paper 'Addressing Data Annotation Challenges in Multiple Sensors: A Solution for Scania Collected Datasets' accepted to ECC 2024; presented in Stockholm.
- Paper 'Semantic 3D grid maps for autonomous driving' accepted to ITSC 2022.
- To present extended abstract 'BlendCLIP: Bridging Synthetic and Real Domains for Zero-Shot 3D Object Classification with Multimodal Pretraining' at OpenSUN3D Workshop in October 2025.
Background
- PhD candidate at the Division of Robotics, Perception and Learning, KTH Royal Institute of Technology, supervised by Prof. Patric Jensfelt and Dr. Sina Sharif Mansouri.
- PhD thesis focuses on long-range perception for autonomous trucking.
- Research interests include scene perception (e.g., object detection, segmentation, scene flow), 3D forecasting, scene rendering, and world models.
- Passionate about developing deployable and scalable methods robust to adverse weather and long-tail scenarios.
- Experienced with various sensors including LiDAR, cameras, and radar.
- Research is part of the ProSense project funded by Vinnova.
