5 papers
Bridging Perspectives: Foundation Model Guided BEV Maps for 3D Object Detection and Tracking
Markus Käppeler, Özgün Çiçek, Daniele Cattaneo +3
Camera-based 3D object detection and tracking are essential for perception in autonomous driving. Current state-of-the-art approaches often rely exclusively on either perspective-v…
Dynamic Robot-Assisted Surgery with Hierarchical Class-Incremental Semantic Segmentation
Julia Hindel, Ema Mekic, Enamundram Naga Karthik +4
Robot-assisted surgeries rely on accurate and real-time scene understanding to safely guide surgical instruments. However, segmentation models trained on static datasets face key l…
Open-Set LiDAR Panoptic Segmentation Guided by Uncertainty-Aware Learning
Rohit Mohan, Julia Hindel, Florian Drews +3
Autonomous vehicles that navigate in open-world environments may encounter previously unseen object classes. However, most existing LiDAR panoptic segmentation models rely on close…
Visual Loop Closure Detection Through Deep Graph Consensus
Martin Büchner, Liza Dahiya, Simon Dorer +4
Visual loop closure detection traditionally relies on place recognition methods to retrieve candidate loops that are validated using computationally expensive RANSAC-based geometri…
Label-Efficient LiDAR Semantic Segmentation with 2D-3D Vision Transformer Adapters
Julia Hindel, Rohit Mohan, Jelena Bratulic +3
LiDAR semantic segmentation models are typically trained from random initialization as universal pre-training is hindered by the lack of large, diverse datasets. Moreover, most poi…