Sequential Recognition of Manipulation Actions Using Discriminative Superpixel Group Mining

Research output: Contribution to conferencePaper

Abstract

Activities such as those involved in food preparation involve interactions between hands, tools and multiple manipulated objects that affect them in visually complex ways making recognition of their constituent actions challenging. We describe a system that classifies action classes in such a setting based on discriminative spatio-temporal superpixel groups. The entire system operates sequentially enabling online action recognition. We obtain state-of-the-art results whilst employing a compact, interpretable representation.
Original languageEnglish
Pages579-583
Number of pages5
DOIs
Publication statusPublished - 6 Sep 2018
Event2018 IEEE International Conference on Image Processing - Athens, Greece
Duration: 7 Oct 201810 Oct 2018
Conference number: 25
https://2018.ieeeicip.org/Default.asp

Conference

Conference2018 IEEE International Conference on Image Processing
Abbreviated titleIEEE ICIP
CountryGreece
CityAthens
Period7/10/1810/10/18
Internet address

Fingerprint

Hand tools

Cite this

Huang, T., & McKenna, S. (2018). Sequential Recognition of Manipulation Actions Using Discriminative Superpixel Group Mining. 579-583. Paper presented at 2018 IEEE International Conference on Image Processing, Athens, Greece. https://doi.org/10.1109/ICIP.2018.8451451
Huang, Tianjun ; McKenna, Stephen. / Sequential Recognition of Manipulation Actions Using Discriminative Superpixel Group Mining. Paper presented at 2018 IEEE International Conference on Image Processing, Athens, Greece.5 p.
@conference{781a96820beb4d0ab1aec4e69e90d58d,
title = "Sequential Recognition of Manipulation Actions Using Discriminative Superpixel Group Mining",
abstract = "Activities such as those involved in food preparation involve interactions between hands, tools and multiple manipulated objects that affect them in visually complex ways making recognition of their constituent actions challenging. We describe a system that classifies action classes in such a setting based on discriminative spatio-temporal superpixel groups. The entire system operates sequentially enabling online action recognition. We obtain state-of-the-art results whilst employing a compact, interpretable representation.",
author = "Tianjun Huang and Stephen McKenna",
note = "This work received funding from the Scottish Informatics & Computer Science Alliance (SICSA) through a prize studentship awarded to T. Huang.; 2018 IEEE International Conference on Image Processing, IEEE ICIP ; Conference date: 07-10-2018 Through 10-10-2018",
year = "2018",
month = "9",
day = "6",
doi = "10.1109/ICIP.2018.8451451",
language = "English",
pages = "579--583",
url = "https://2018.ieeeicip.org/Default.asp",

}

Huang, T & McKenna, S 2018, 'Sequential Recognition of Manipulation Actions Using Discriminative Superpixel Group Mining' Paper presented at 2018 IEEE International Conference on Image Processing, Athens, Greece, 7/10/18 - 10/10/18, pp. 579-583. https://doi.org/10.1109/ICIP.2018.8451451

Sequential Recognition of Manipulation Actions Using Discriminative Superpixel Group Mining. / Huang, Tianjun; McKenna, Stephen.

2018. 579-583 Paper presented at 2018 IEEE International Conference on Image Processing, Athens, Greece.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Sequential Recognition of Manipulation Actions Using Discriminative Superpixel Group Mining

AU - Huang, Tianjun

AU - McKenna, Stephen

N1 - This work received funding from the Scottish Informatics & Computer Science Alliance (SICSA) through a prize studentship awarded to T. Huang.

PY - 2018/9/6

Y1 - 2018/9/6

N2 - Activities such as those involved in food preparation involve interactions between hands, tools and multiple manipulated objects that affect them in visually complex ways making recognition of their constituent actions challenging. We describe a system that classifies action classes in such a setting based on discriminative spatio-temporal superpixel groups. The entire system operates sequentially enabling online action recognition. We obtain state-of-the-art results whilst employing a compact, interpretable representation.

AB - Activities such as those involved in food preparation involve interactions between hands, tools and multiple manipulated objects that affect them in visually complex ways making recognition of their constituent actions challenging. We describe a system that classifies action classes in such a setting based on discriminative spatio-temporal superpixel groups. The entire system operates sequentially enabling online action recognition. We obtain state-of-the-art results whilst employing a compact, interpretable representation.

U2 - 10.1109/ICIP.2018.8451451

DO - 10.1109/ICIP.2018.8451451

M3 - Paper

SP - 579

EP - 583

ER -

Huang T, McKenna S. Sequential Recognition of Manipulation Actions Using Discriminative Superpixel Group Mining. 2018. Paper presented at 2018 IEEE International Conference on Image Processing, Athens, Greece. https://doi.org/10.1109/ICIP.2018.8451451