Thesis icon

Thesis

Object-centric generative models for robot perception and action

Abstract:

The system of robot manipulation involves a pipeline consisting of the perception of objects in the environment and the planning of actions in 3D space. Deep learning approaches are employed to segment scenes into components of objects and then learn object-centric features to predict actions for downstream tasks. Despite having achieved promising performance in several manipulation tasks, supervised approaches lack inductive biases related to general properties of objects. Recent advances...

Expand abstract

Actions

Access Document

Files:

Authors

More by this author
Institution:
University of Oxford
Division:
MPLS
Department:
Engineering Science
Role:
Author

Contributors

Institution:
University of Oxford
Division:
MPLS
Department:
Engineering Science
Role:
Supervisor
Institution:
University of Oxford
Division:
MPLS
Department:
Computer Science
Role:
Examiner
Role:
Examiner


DOI:
Type of award:
DPhil
Level of award:
Doctoral
Awarding institution:
University of Oxford

Terms of use


Views and Downloads






If you are the owner of this record, you can report an update to it here: Report update to this record

TO TOP