Abstract
Manipulation action recognition is a challenging problem in computer vision. We previously reported a system based on matching groups of superpixels. In this paper, we modify the superpixel group mining algorithm and report results on two datasets. Recognition accuracies are comparable with those reported using deep learning. The representation used in our approach is amenable to interpretation. Specifically, visualisation of matched groups provides a level of explanation for recognition decisions and insights into the likely generalisation ability of action representations.
Original language | English |
---|---|
Title of host publication | ReaLX 2018 |
Subtitle of host publication | Proceedings of the SICSA Workshop on Reasoning, Learning and Explainability |
Editors | Kyle Martin , Nirmalie Wiratunga, Leslie S. Smith |
Publisher | CEUR-WS |
Pages | 1-5 |
Number of pages | 5 |
Volume | 2151 |
Publication status | Published - 27 Jul 2018 |
Publication series
Name | CEUR Workshop Proceedings |
---|---|
Publisher | CEUR |
ISSN (Electronic) | 1613-0073 |
Keywords
- Action recognition
- Computer vision
- Superpixel group mining
ASJC Scopus subject areas
- General Computer Science
Fingerprint
Dive into the research topics of 'Superpixel Group Mining for Manipulation Action Recognition'. Together they form a unique fingerprint.Student theses
-
Sequential Recognition of Manipulation Actions Using Superpixel Group Mining
Huang, T. (Author), McKenna, S. (Supervisor) & Zhang, J. (Supervisor), 2019Student thesis: Doctoral Thesis › Doctor of Philosophy
File