Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
Hamburg University of Technology, Institute of Aircraft Production Technology Corresponding author: keno.moenck@tuhh.de Abstract Autonomous vision applications in production, intralogistics, or manufacturing environments require perception capabilities beyond a small, fixed set of classes. Recent open-vocabulary methods, leveraging
1
0
3
Pre-trained open-vocabulary segmentation models stumble in industrial settings, but a training-free superpoint merging pipeline with a domain-adapted VLFM can effectively segment industrial objects.