3 papers
cs.CV2026
VoxelFix: Post-Hoc Semantic Correction of Completed 3D Voxel Maps
Sunesh Praveen Raja Sundarasami, Taehyoung Kim, Johannes Scherer +5
Semantic 3D maps are increasingly constructed automatically for aerial robotics by integrating learned semantic predictions into 3D representations. While this avoids costly manual…
cs.RO2025
Parameter-Efficient Fine-Tuning of Vision Foundation Model for Forest Floor Segmentation from UAV Imagery
Mohammad Wasil, Ahmad Drak, Brennan Penfold +4
Unmanned Aerial Vehicles (UAVs) are increasingly used for reforestation and forest monitoring, including seed dispersal in hard-to-reach terrains. However, a detailed understanding…
cs.CV2025
Taming the Randomness: Towards Label-Preserving Cropping in Contrastive Learning
Mohamed Hassan, Mohammad Wasil, Sebastian Houben
Contrastive learning (CL) approaches have gained great recognition as a very successful subset of self-supervised learning (SSL) methods. SSL enables learning from unlabeled data,…