Canonical mapping as a general purpose representation
Author(s)
Rmiche, Nabil
Advisor(s)
Editor(s)
Collections
Supplementary to:
Permanent Link
Abstract
Perception is critical for robotic manipulation in open environments, where traditional approaches often produce task-specific predictions that are unsuitable for deformable objects or adaptation to other tasks. In this work, conducted within a USDA-funded initiative focused on automating the poultry processing industry, we propose canonical mapping as a near-universal and flexible object descriptor. Canonical mapping establishes correspondences between image pixels and a 3D mesh, enabling robust pose prediction for both rigid and deformable objects.
Sponsor
Date
2024-12-12
Extent
Resource Type
Text
Resource Subtype
Thesis (Masters Degree)