Action Recognition From Skeleton Data via Analogical Generalization Over Qualitative Representations

Kezhen Chen, Kenneth D Forbus

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

recognition accuracy, but they are typically black boxes whose internal models are not inspectable and whose results are not explainable. This paper describes a new pipeline for recognizing human actions from skeleton data via analogical generalization. Specifically, starting with Kinect data, we segment each human action by temporal regions where the motion is qualitatively uniform, creating a sketch graph that provides a form of qualitative representation of the behavior that is easy to visualize. Models are learned from sketch graphs via analogical generalization, which are then used for classification via analogical retrieval. The retrieval process also produces links between the new example and components of the model that provide explanations. To improve recognition accuracy, we implement dynamic feature selection to pick reasonable relational features. We show the explanation advantage of our approach by example, and results on three public datasets illustrate its utility.
Original languageEnglish (US)
Title of host publicationProceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, 2018
EditorsSheila A McIlraith, Kilian Q Weinberger
PublisherAAAI Press
StatePublished - 2018

Fingerprint

Feature extraction
Pipelines

Cite this

Chen, K., & Forbus, K. D. (2018). Action Recognition From Skeleton Data via Analogical Generalization Over Qualitative Representations. In S. A. McIlraith, & K. Q. Weinberger (Eds.), Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, 2018 AAAI Press.
Chen, Kezhen ; Forbus, Kenneth D. / Action Recognition From Skeleton Data via Analogical Generalization Over Qualitative Representations. Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, 2018. editor / Sheila A McIlraith ; Kilian Q Weinberger. AAAI Press, 2018.
@inproceedings{f93433dde8ad4f78a1dd4f229b571c8e,
title = "Action Recognition From Skeleton Data via Analogical Generalization Over Qualitative Representations",
abstract = "recognition accuracy, but they are typically black boxes whose internal models are not inspectable and whose results are not explainable. This paper describes a new pipeline for recognizing human actions from skeleton data via analogical generalization. Specifically, starting with Kinect data, we segment each human action by temporal regions where the motion is qualitatively uniform, creating a sketch graph that provides a form of qualitative representation of the behavior that is easy to visualize. Models are learned from sketch graphs via analogical generalization, which are then used for classification via analogical retrieval. The retrieval process also produces links between the new example and components of the model that provide explanations. To improve recognition accuracy, we implement dynamic feature selection to pick reasonable relational features. We show the explanation advantage of our approach by example, and results on three public datasets illustrate its utility.",
author = "Kezhen Chen and Forbus, {Kenneth D}",
note = "https://www.aaai.org/ocs/index.php/AAAI/AAAI18/paper/view/16616",
year = "2018",
language = "English (US)",
editor = "McIlraith, {Sheila A} and Weinberger, {Kilian Q}",
booktitle = "Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, 2018",
publisher = "AAAI Press",

}

Chen, K & Forbus, KD 2018, Action Recognition From Skeleton Data via Analogical Generalization Over Qualitative Representations. in SA McIlraith & KQ Weinberger (eds), Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, 2018. AAAI Press.

Action Recognition From Skeleton Data via Analogical Generalization Over Qualitative Representations. / Chen, Kezhen; Forbus, Kenneth D.

Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, 2018. ed. / Sheila A McIlraith; Kilian Q Weinberger. AAAI Press, 2018.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Action Recognition From Skeleton Data via Analogical Generalization Over Qualitative Representations

AU - Chen, Kezhen

AU - Forbus, Kenneth D

N1 - https://www.aaai.org/ocs/index.php/AAAI/AAAI18/paper/view/16616

PY - 2018

Y1 - 2018

N2 - recognition accuracy, but they are typically black boxes whose internal models are not inspectable and whose results are not explainable. This paper describes a new pipeline for recognizing human actions from skeleton data via analogical generalization. Specifically, starting with Kinect data, we segment each human action by temporal regions where the motion is qualitatively uniform, creating a sketch graph that provides a form of qualitative representation of the behavior that is easy to visualize. Models are learned from sketch graphs via analogical generalization, which are then used for classification via analogical retrieval. The retrieval process also produces links between the new example and components of the model that provide explanations. To improve recognition accuracy, we implement dynamic feature selection to pick reasonable relational features. We show the explanation advantage of our approach by example, and results on three public datasets illustrate its utility.

AB - recognition accuracy, but they are typically black boxes whose internal models are not inspectable and whose results are not explainable. This paper describes a new pipeline for recognizing human actions from skeleton data via analogical generalization. Specifically, starting with Kinect data, we segment each human action by temporal regions where the motion is qualitatively uniform, creating a sketch graph that provides a form of qualitative representation of the behavior that is easy to visualize. Models are learned from sketch graphs via analogical generalization, which are then used for classification via analogical retrieval. The retrieval process also produces links between the new example and components of the model that provide explanations. To improve recognition accuracy, we implement dynamic feature selection to pick reasonable relational features. We show the explanation advantage of our approach by example, and results on three public datasets illustrate its utility.

M3 - Conference contribution

BT - Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, 2018

A2 - McIlraith, Sheila A

A2 - Weinberger, Kilian Q

PB - AAAI Press

ER -

Chen K, Forbus KD. Action Recognition From Skeleton Data via Analogical Generalization Over Qualitative Representations. In McIlraith SA, Weinberger KQ, editors, Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, 2018. AAAI Press. 2018