10 papers
Per-Group Error, Not Total MSE: Fine-Tuning Vision-Language-Action Models for 11-DoF Mobile Manipulation
Pau Montagut Bofi, Mario GarcÃa Blasco, Tessa Pulli +1
Fine-tuning Vision-Language-Action (VLA) models for mobile manipulators with heterogeneous joint spaces can produce a counterintuitive result: the checkpoint with the lowest aggreg…
OSCAR: Open-Set CAD Retrieval from a Language Prompt and a Single Image
Tessa Pulli, Jean-Baptiste Weibel, Peter Hönig +3
6D object pose estimation plays a crucial role in scene understanding for applications such as robotics and augmented reality. To support the needs of ever-changing object sets in…
SCOPE: Semantic Conditioning for Sim2Real Category-Level Object Pose Estimation in Robotics
Peter Hönig, Peter Hönig, Stefan Thalhammer +3
Object manipulation requires accurate object pose estimation. In open environments, robots encounter unknown objects, which requires semantic understanding in order to generalize b…
Sim2Real Transfer for Vision-Based Grasp Verification
Pau Amargant, Peter Hönig, Markus Vincze
The verification of successful grasps is a crucial aspect of robot manipulation, particularly when handling deformable objects. Traditional methods relying on force and tactile sen…
LLM-Empowered Embodied Agent for Memory-Augmented Task Planning in Household Robotics
Marc Glocker, Peter Hönig, Matthias Hirschmanner +1
We present an embodied robotic system with an LLM-driven agent-orchestration architecture for autonomous household object management. The system integrates memory-augmented task pl…
Category-Level and Open-Set Object Pose Estimation for Robotics
Peter Hönig, Matthias Hirschmanner, Markus Vincze
Object pose estimation enables a variety of tasks in computer vision and robotics, including scene understanding and robotic grasping. The complexity of a pose estimation task depe…