Beyond Action Recognition: Action Completion in RGB-D Data

Farnoosh Heidarivincheh, Majid Mirmehdi and Dima Damen

Abstract

An action is completed when its goal has been successfully achieved. Using current state-of-the-art depth features, designed primarily for action recognition, an incomplete sequence may still be classified as its complete counterpart due to the overlap in evidence. In this work we show that while features can perform comparably for action recognition, they vary in their ability to recognise incompletion. Experimenting on a novel dataset of 414 complete/incomplete object interaction sequences, spanning six actions and captured using an RGB-D camera, we test for completion using binary classification on labelled data. Results show that by selecting the suitable feature per action, we achieve 95.7% accuracy for recognising action completion.

Session

Video events, robot vision and deep learning

Files

Extended Abstract (PDF, 1M)

Paper (PDF, 4M)

Supplemental Materials (ZIP, 28M)

DOI

10.5244/C.30.142
https://dx.doi.org/10.5244/C.30.142

Citation

Farnoosh Heidarivincheh, Majid Mirmehdi and Dima Damen. Beyond Action Recognition: Action Completion in RGB-D Data. In Richard C. Wilson, Edwin R. Hancock and William A. P. Smith, editors, Proceedings of the British Machine Vision Conference (BMVC), pages 142.1-142.11. BMVA Press, September 2016.

Bibtex

        @inproceedings{BMVC2016_142,
        	title={Beyond Action Recognition: Action Completion in RGB-D Data},
        	author={Farnoosh Heidarivincheh, Majid Mirmehdi and Dima Damen},
        	year={2016},
        	month={September},
        	pages={142.1-142.11},
        	articleno={142},
        	numpages={11},
        	booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
        	publisher={BMVA Press},
        	editor={Richard C. Wilson, Edwin R. Hancock and William A. P. Smith},
        	doi={10.5244/C.30.142},
        	isbn={1-901725-59-6},
        	url={https://dx.doi.org/10.5244/C.30.142}
        }