Towards an Unequivocal Representation of Actions

Michael Wray, Davide Moltisanti, Dima Damen; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2018, pp. 1127-1131

Abstract


This work introduces verb-only representations for actions and interactions; the problem of describing similar motions (e.g. `open door', `open cupboard'), and distinguish differing ones (e.g. `open door' vs `open bottle') using verb-only labels. Current approaches neglect legitimate semantic ambiguities and class overlaps between verbs (Fig. 1), relying on the objects to disambiguate interactions. We deviate from single-verb labels and introduce a mapping between observations and multiple verb labels -- in order to create an Unequivocal Representation of Actions. The new representation benefits from increased vocabulary and a soft assignment to an enriched space of verb labels. We learn these representations as multi-output regression, using a two-stream fusion CNN. The proposed approach outperforms conventional single-verb labels (also known as majority voting) on three egocentric datasets for both recognition and retrieval.

Related Material


[pdf] [arXiv]
[bibtex]
@InProceedings{Wray_2018_CVPR_Workshops,
author = {Wray, Michael and Moltisanti, Davide and Damen, Dima},
title = {Towards an Unequivocal Representation of Actions},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2018}
}