Abstract

Gesture spotting is the task of detecting and recognizing gestures defined in a vocabulary. The difficulty of gesture spotting stems from the fact that valid gestures appear sporadically in a continuous gesture stream, interspersed with invalid gestures (movements that do not correspond to any gesture contained in the vocabulary). In this paper, a novel method for designing threshold models from valid gesture models learnt through Adaptive Boosting is proposed. This threshold model is adaptive in nature and discriminates between valid and invalid gestures. Furthermore, a gesture spotting network consisting of the individual gesture models and the threshold model is proposed to perform the task of spotting and recognition simultaneously. This technique is evaluated in the context of spotting and recognizing activity gestures (hand gestures) from continuous accelerometer data streams. The proposed technique results in a precision of 0.78 and a recall of 0.93 out performing the HMM based threshold model which resulted in 0.4 and 0.81 precision and recall values.

Original languageEnglish (US)
Title of host publication2010 IEEE International Conference on Multimedia and Expo, ICME 2010
Pages155-160
Number of pages6
DOIs
StatePublished - 2010
Event2010 IEEE International Conference on Multimedia and Expo, ICME 2010 - Singapore, Singapore
Duration: Jul 19 2010Jul 23 2010

Other

Other2010 IEEE International Conference on Multimedia and Expo, ICME 2010
CountrySingapore
CitySingapore
Period7/19/107/23/10

Fingerprint

Adaptive boosting
Accelerometers

Keywords

  • Accelerometer
  • Activity gestures
  • Adaptive boosting
  • Gesture spotting
  • Viterbi algorithm

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Software

Cite this

Krishnan, N. C., Lade, P., & Panchanathan, S. (2010). Activity gesture spotting using a threshold model based on adaptive boosting. In 2010 IEEE International Conference on Multimedia and Expo, ICME 2010 (pp. 155-160). [5583013] https://doi.org/10.1109/ICME.2010.5583013

Activity gesture spotting using a threshold model based on adaptive boosting. / Krishnan, Narayanan C.; Lade, Prasanth; Panchanathan, Sethuraman.

2010 IEEE International Conference on Multimedia and Expo, ICME 2010. 2010. p. 155-160 5583013.

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

Krishnan, NC, Lade, P & Panchanathan, S 2010, Activity gesture spotting using a threshold model based on adaptive boosting. in 2010 IEEE International Conference on Multimedia and Expo, ICME 2010., 5583013, pp. 155-160, 2010 IEEE International Conference on Multimedia and Expo, ICME 2010, Singapore, Singapore, 7/19/10. https://doi.org/10.1109/ICME.2010.5583013
Krishnan NC, Lade P, Panchanathan S. Activity gesture spotting using a threshold model based on adaptive boosting. In 2010 IEEE International Conference on Multimedia and Expo, ICME 2010. 2010. p. 155-160. 5583013 https://doi.org/10.1109/ICME.2010.5583013
Krishnan, Narayanan C. ; Lade, Prasanth ; Panchanathan, Sethuraman. / Activity gesture spotting using a threshold model based on adaptive boosting. 2010 IEEE International Conference on Multimedia and Expo, ICME 2010. 2010. pp. 155-160
@inproceedings{d3ee6a8a8fec49ed98fef0bf507965a4,
title = "Activity gesture spotting using a threshold model based on adaptive boosting",
abstract = "Gesture spotting is the task of detecting and recognizing gestures defined in a vocabulary. The difficulty of gesture spotting stems from the fact that valid gestures appear sporadically in a continuous gesture stream, interspersed with invalid gestures (movements that do not correspond to any gesture contained in the vocabulary). In this paper, a novel method for designing threshold models from valid gesture models learnt through Adaptive Boosting is proposed. This threshold model is adaptive in nature and discriminates between valid and invalid gestures. Furthermore, a gesture spotting network consisting of the individual gesture models and the threshold model is proposed to perform the task of spotting and recognition simultaneously. This technique is evaluated in the context of spotting and recognizing activity gestures (hand gestures) from continuous accelerometer data streams. The proposed technique results in a precision of 0.78 and a recall of 0.93 out performing the HMM based threshold model which resulted in 0.4 and 0.81 precision and recall values.",
keywords = "Accelerometer, Activity gestures, Adaptive boosting, Gesture spotting, Viterbi algorithm",
author = "Krishnan, {Narayanan C.} and Prasanth Lade and Sethuraman Panchanathan",
year = "2010",
doi = "10.1109/ICME.2010.5583013",
language = "English (US)",
isbn = "9781424474912",
pages = "155--160",
booktitle = "2010 IEEE International Conference on Multimedia and Expo, ICME 2010",

}

TY - GEN

T1 - Activity gesture spotting using a threshold model based on adaptive boosting

AU - Krishnan, Narayanan C.

AU - Lade, Prasanth

AU - Panchanathan, Sethuraman

PY - 2010

Y1 - 2010

N2 - Gesture spotting is the task of detecting and recognizing gestures defined in a vocabulary. The difficulty of gesture spotting stems from the fact that valid gestures appear sporadically in a continuous gesture stream, interspersed with invalid gestures (movements that do not correspond to any gesture contained in the vocabulary). In this paper, a novel method for designing threshold models from valid gesture models learnt through Adaptive Boosting is proposed. This threshold model is adaptive in nature and discriminates between valid and invalid gestures. Furthermore, a gesture spotting network consisting of the individual gesture models and the threshold model is proposed to perform the task of spotting and recognition simultaneously. This technique is evaluated in the context of spotting and recognizing activity gestures (hand gestures) from continuous accelerometer data streams. The proposed technique results in a precision of 0.78 and a recall of 0.93 out performing the HMM based threshold model which resulted in 0.4 and 0.81 precision and recall values.

AB - Gesture spotting is the task of detecting and recognizing gestures defined in a vocabulary. The difficulty of gesture spotting stems from the fact that valid gestures appear sporadically in a continuous gesture stream, interspersed with invalid gestures (movements that do not correspond to any gesture contained in the vocabulary). In this paper, a novel method for designing threshold models from valid gesture models learnt through Adaptive Boosting is proposed. This threshold model is adaptive in nature and discriminates between valid and invalid gestures. Furthermore, a gesture spotting network consisting of the individual gesture models and the threshold model is proposed to perform the task of spotting and recognition simultaneously. This technique is evaluated in the context of spotting and recognizing activity gestures (hand gestures) from continuous accelerometer data streams. The proposed technique results in a precision of 0.78 and a recall of 0.93 out performing the HMM based threshold model which resulted in 0.4 and 0.81 precision and recall values.

KW - Accelerometer

KW - Activity gestures

KW - Adaptive boosting

KW - Gesture spotting

KW - Viterbi algorithm

UR - http://www.scopus.com/inward/record.url?scp=78349289224&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78349289224&partnerID=8YFLogxK

U2 - 10.1109/ICME.2010.5583013

DO - 10.1109/ICME.2010.5583013

M3 - Conference contribution

AN - SCOPUS:78349289224

SN - 9781424474912

SP - 155

EP - 160

BT - 2010 IEEE International Conference on Multimedia and Expo, ICME 2010

ER -