Papers
101
Total Citations
7,338
H-Index
30
About
Sebastian Scherer is a pioneering robotics researcher whose work spans autonomous navigation, 3D perception, and simultaneous localization and mapping (SLAM). Perhaps his most influential contribution is co-authoring VoxNet (2015), a landmark paper introducing 3D convolutional neural networks for real-time object recognition using LiDAR and RGBD data, which has amassed an extraordinary 3,579 citations and helped define modern robotic perception. His research consistently addresses the challenge of robust robot operation in complex, degraded environments — from underground tunnels to underwater systems — as demonstrated by his contributions to the UUV Simulator, Super Odometry, and the highly regarded DARPA SubT Challenge survey. His TartanAir dataset (365 citations) pushed the boundaries of visual SLAM research by leveraging photorealistic simulation, while SplaTAM (2024, 323 citations) introduced Gaussian splatting to dense RGB-D SLAM with immediate research impact. Scherer has also advanced UAV safety through fault detection datasets and explored deep reinforcement learning for continuous control. With over 5,500 cumulative citations across diverse domains, his work is essential reading for researchers building reliable, intelligent autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1VoxNet: A 3D Convolutional Neural Network for real-time object recognition3,579 citations · 2015
- 2
- 3TartanAir: A Dataset to Push the Limits of Visual SLAM365 citations · 2020
- 4SplaTAM: Splat, Track & Map 3D Gaussians for Dense RGB-D SLAM323 citations · 2024
- 5
- 6Present and Future of SLAM in Extreme Environments: The DARPA SubT Challenge176 citations · 2023
- 7<i>AnyLoc</i>: Towards Universal Visual Place Recognition163 citations · 2023
- 8
- 9ALFA: A dataset for UAV fault and anomaly detection123 citations · 2020
- 10