Perception-Action Based Object Detection From Local Descriptor Combination And Reinforcement Learning

Lucas Paletta; Gerald Fritz; Christin Seifert

doi:10.1007/11499145_65

Perception-Action Based Object Detection From Local Descriptor Combination And Reinforcement Learning

Lucas Paletta, Gerald Fritz, Christin Seifert

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Abstract

This work proposes to learn visual encodings of attention patterns\nthat enables sequential attention for object detection in real world\nenvironments. The system embeds a saccadic decision procedure in\na cascaded process where visual evidence is probed at informative\nimage locations. It is based on the extraction of information theoretic\nsaliency by determining informative local image descriptors that\nprovide selected foci of interest. The local information in terms\nof code book vector responses and the geometric information in the\nshift of attention contribute to recognition states of a Markov decision\nprocess. A Q-learner performs then performs search on useful actions\ntowards salient locations, developing a strategy of action sequences\ndirected in state space towards the optimization of information maximization.\nThe method is evaluated in outdoor object recognition and demonstrates\nefficient performance.

Original language	English
Title of host publication	Image Analysis
Subtitle of host publication	14th Scandinavian Conference, SCIA 2005, Joensuu, Finland, June 19-22, 2005. Proceedings
Editors	Heikki Kalviainen, Jussi Parkkinen, Arto Kaarna
Publisher	Springer
Pages	639-648
Number of pages	10
ISBN (Electronic)	978-3-540-31566-7
ISBN (Print)	978-3-540-26320-3
DOIs	https://doi.org/10.1007/11499145_65
Publication status	Published - 2005
Event	14th Scandinavian Conference on Image Analysis, SCIA 2005 - Joensuu, Finland Duration: 19 Jun 2005 → 22 Jun 2005 Conference number: 14

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	3540
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	14th Scandinavian Conference on Image Analysis, SCIA 2005
Abbreviated title	SCIA
Country/Territory	Finland
City	Joensuu
Period	19/06/05 → 22/06/05

Access to Document

10.1007/11499145_65

Cite this

Paletta, L., Fritz, G., & Seifert, C. (2005). Perception-Action Based Object Detection From Local Descriptor Combination And Reinforcement Learning. In H. Kalviainen, J. Parkkinen, & A. Kaarna (Eds.), Image Analysis: 14th Scandinavian Conference, SCIA 2005, Joensuu, Finland, June 19-22, 2005. Proceedings (pp. 639-648). (Lecture Notes in Computer Science; Vol. 3540). Springer. https://doi.org/10.1007/11499145_65

Paletta, Lucas ; Fritz, Gerald ; Seifert, Christin. / Perception-Action Based Object Detection From Local Descriptor Combination And Reinforcement Learning. Image Analysis: 14th Scandinavian Conference, SCIA 2005, Joensuu, Finland, June 19-22, 2005. Proceedings. editor / Heikki Kalviainen ; Jussi Parkkinen ; Arto Kaarna. Springer, 2005. pp. 639-648 (Lecture Notes in Computer Science).

@inproceedings{ea417aae0d2d41e180ee5b03fd3c845a,

title = "Perception-Action Based Object Detection From Local Descriptor Combination And Reinforcement Learning",

abstract = "This work proposes to learn visual encodings of attention patterns\nthat enables sequential attention for object detection in real world\nenvironments. The system embeds a saccadic decision procedure in\na cascaded process where visual evidence is probed at informative\nimage locations. It is based on the extraction of information theoretic\nsaliency by determining informative local image descriptors that\nprovide selected foci of interest. The local information in terms\nof code book vector responses and the geometric information in the\nshift of attention contribute to recognition states of a Markov decision\nprocess. A Q-learner performs then performs search on useful actions\ntowards salient locations, developing a strategy of action sequences\ndirected in state space towards the optimization of information maximization.\nThe method is evaluated in outdoor object recognition and demonstrates\nefficient performance.",

author = "Lucas Paletta and Gerald Fritz and Christin Seifert",

year = "2005",

doi = "10.1007/11499145_65",

language = "English",

isbn = "978-3-540-26320-3",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "639--648",

editor = "Heikki Kalviainen and Jussi Parkkinen and Arto Kaarna",

booktitle = "Image Analysis",

address = "Netherlands",

note = "14th Scandinavian Conference on Image Analysis, SCIA 2005, SCIA ; Conference date: 19-06-2005 Through 22-06-2005",

}

Paletta, L, Fritz, G & Seifert, C 2005, Perception-Action Based Object Detection From Local Descriptor Combination And Reinforcement Learning. in H Kalviainen, J Parkkinen & A Kaarna (eds), Image Analysis: 14th Scandinavian Conference, SCIA 2005, Joensuu, Finland, June 19-22, 2005. Proceedings. Lecture Notes in Computer Science, vol. 3540, Springer, pp. 639-648, 14th Scandinavian Conference on Image Analysis, SCIA 2005, Joensuu, Finland, 19/06/05. https://doi.org/10.1007/11499145_65

Perception-Action Based Object Detection From Local Descriptor Combination And Reinforcement Learning. / Paletta, Lucas; Fritz, Gerald; Seifert, Christin.

Image Analysis: 14th Scandinavian Conference, SCIA 2005, Joensuu, Finland, June 19-22, 2005. Proceedings. ed. / Heikki Kalviainen; Jussi Parkkinen; Arto Kaarna. Springer, 2005. p. 639-648 (Lecture Notes in Computer Science; Vol. 3540).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Perception-Action Based Object Detection From Local Descriptor Combination And Reinforcement Learning

AU - Paletta, Lucas

AU - Fritz, Gerald

AU - Seifert, Christin

N1 - Conference code: 14

PY - 2005

Y1 - 2005

N2 - This work proposes to learn visual encodings of attention patterns\nthat enables sequential attention for object detection in real world\nenvironments. The system embeds a saccadic decision procedure in\na cascaded process where visual evidence is probed at informative\nimage locations. It is based on the extraction of information theoretic\nsaliency by determining informative local image descriptors that\nprovide selected foci of interest. The local information in terms\nof code book vector responses and the geometric information in the\nshift of attention contribute to recognition states of a Markov decision\nprocess. A Q-learner performs then performs search on useful actions\ntowards salient locations, developing a strategy of action sequences\ndirected in state space towards the optimization of information maximization.\nThe method is evaluated in outdoor object recognition and demonstrates\nefficient performance.

AB - This work proposes to learn visual encodings of attention patterns\nthat enables sequential attention for object detection in real world\nenvironments. The system embeds a saccadic decision procedure in\na cascaded process where visual evidence is probed at informative\nimage locations. It is based on the extraction of information theoretic\nsaliency by determining informative local image descriptors that\nprovide selected foci of interest. The local information in terms\nof code book vector responses and the geometric information in the\nshift of attention contribute to recognition states of a Markov decision\nprocess. A Q-learner performs then performs search on useful actions\ntowards salient locations, developing a strategy of action sequences\ndirected in state space towards the optimization of information maximization.\nThe method is evaluated in outdoor object recognition and demonstrates\nefficient performance.

UR - http://www.mendeley.com/research/perceptionaction-based-object-detection-local-descriptor-combination-reinforcement-learning

U2 - 10.1007/11499145_65

DO - 10.1007/11499145_65

M3 - Conference contribution

SN - 978-3-540-26320-3

T3 - Lecture Notes in Computer Science

SP - 639

EP - 648

BT - Image Analysis

A2 - Kalviainen, Heikki

A2 - Parkkinen, Jussi

A2 - Kaarna, Arto

PB - Springer

T2 - 14th Scandinavian Conference on Image Analysis, SCIA 2005

Y2 - 19 June 2005 through 22 June 2005

ER -

Paletta L, Fritz G, Seifert C. Perception-Action Based Object Detection From Local Descriptor Combination And Reinforcement Learning. In Kalviainen H, Parkkinen J, Kaarna A, editors, Image Analysis: 14th Scandinavian Conference, SCIA 2005, Joensuu, Finland, June 19-22, 2005. Proceedings. Springer. 2005. p. 639-648. (Lecture Notes in Computer Science). https://doi.org/10.1007/11499145_65

Perception-Action Based Object Detection From Local Descriptor Combination And Reinforcement Learning

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this