no code implementations • 19 Sep 2020 • Nils Keunecke, S. Hamidreza Kasaei
Therefore, robots should have the functionality to learn about new object categories in an open-ended fashion while working in the environment. Towards this goal, we propose a deep transfer learning approach to generate a scale- and pose-invariant object representation by considering shape and texture information in multiple colorspaces.
no code implementations • 18 Mar 2020 • S. Hamidreza Kasaei, Jorik Melsen, Floris van Beers, Christiaan Steenkist, Klemen Voncina
In this way, the robot will constantly learn how to help humans in everyday tasks by gaining more and more experiences without the need for re-programming.
no code implementations • 10 Feb 2020 • S. Hamidreza Kasaei, Maryam Ghorbani, Jits Schilperoort, Wessel van der Rest
In this paper, we explore the importance of shape information, color constancy, color spaces, and various similarity measures in open-ended 3D object recognition.
1 code implementation • 10 Feb 2020 • Yikun Li, Lambert Schomaker, S. Hamidreza Kasaei
Affordance detection is one of the challenging tasks in robotics because it must predict the grasp configuration for the object of interest in real-time to enable the robot to interact with the environment.
Robotics
no code implementations • 19 Dec 2019 • S. Hamidreza Kasaei
In particular, this architecture provides perception capabilities that will allow robots to incrementally learn object categories from the set of accumulated experiences and reason about how to perform complex tasks.
no code implementations • 26 Jul 2019 • S. Hamidreza Kasaei
In this work, each object is represented using a set of general latent visual topics and category-specific dictionaries.
no code implementations • 25 Jul 2019 • S. Hamidreza Kasaei
Open-ended learning is one of these challenges waiting for many improvements.
Robotics