User independent gesture interaction for small handheld devices

Sanna Kallio, Juha Kela, Panu Korpipää, Jani Mäntyjärvi

Research output: Contribution to journalArticleScientificpeer-review

32 Citations (Scopus)

Abstract

Accelerometer-based gesture recognition facilitates a complementary interaction modality for controlling mobile devices and home appliances. Using gestures for the task of home appliance control requires use of the same device and gestures by different persons, i.e. user independent gesture recognition. The practical application in small embedded low-resource devices also requires high computational performance. The user independent gesture recognition accuracy was evaluated with a set of eight gestures and seven users, with a total of 1120 gestures in the dataset. Twenty-state continuous HMM yielded an average of 96.9% user independent recognition accuracy, which was cross-validated by leaving one user in turn out of the training set. Continuous and discrete five-state HMM computational performances were compared with a reference test in a PC environment, indicating that discrete HMM is 20% faster. Computational performance of discrete five-state HMM was evaluated in an embedded hardware environment with a 104 MHz ARM-9 processor and Symbian OS. The average recognition time per gesture calculated from 1120 gesture repetitions was 8.3 ms. With this result, the computational performance difference between the compared methods is considered insignificant in terms of practical application. Continuous HMM is hence recommended as a preferred method due to its better suitability for a continuous-valued signal, and better recognition accuracy. The results suggest that, according to both evaluation criteria, HMM is feasible for practical user independent gesture control applications in mobile low-resource embedded environments.
Original languageEnglish
Pages (from-to)505-524
Number of pages20
JournalInternational Journal of Pattern Recognition and Artificial Intelligence
Volume20
Issue number4
DOIs
Publication statusPublished - 2006
MoE publication typeA1 Journal article-refereed

Fingerprint

Gesture recognition
Domestic appliances
Mobile homes
Accelerometers
Mobile devices
Hardware

Keywords

  • human computer interaction
  • input technology
  • mobile devices
  • gesture recognition
  • gesture control
  • user interfaces
  • user interaction technology
  • pointing
  • multimodal interaction
  • accelerometer

Cite this

@article{06700a3d68244e2696787d2994515f98,
title = "User independent gesture interaction for small handheld devices",
abstract = "Accelerometer-based gesture recognition facilitates a complementary interaction modality for controlling mobile devices and home appliances. Using gestures for the task of home appliance control requires use of the same device and gestures by different persons, i.e. user independent gesture recognition. The practical application in small embedded low-resource devices also requires high computational performance. The user independent gesture recognition accuracy was evaluated with a set of eight gestures and seven users, with a total of 1120 gestures in the dataset. Twenty-state continuous HMM yielded an average of 96.9{\%} user independent recognition accuracy, which was cross-validated by leaving one user in turn out of the training set. Continuous and discrete five-state HMM computational performances were compared with a reference test in a PC environment, indicating that discrete HMM is 20{\%} faster. Computational performance of discrete five-state HMM was evaluated in an embedded hardware environment with a 104 MHz ARM-9 processor and Symbian OS. The average recognition time per gesture calculated from 1120 gesture repetitions was 8.3 ms. With this result, the computational performance difference between the compared methods is considered insignificant in terms of practical application. Continuous HMM is hence recommended as a preferred method due to its better suitability for a continuous-valued signal, and better recognition accuracy. The results suggest that, according to both evaluation criteria, HMM is feasible for practical user independent gesture control applications in mobile low-resource embedded environments.",
keywords = "human computer interaction, input technology, mobile devices, gesture recognition, gesture control, user interfaces, user interaction technology, pointing, multimodal interaction, accelerometer",
author = "Sanna Kallio and Juha Kela and Panu Korpip{\"a}{\"a} and Jani M{\"a}ntyj{\"a}rvi",
year = "2006",
doi = "10.1142/S0218001406004776",
language = "English",
volume = "20",
pages = "505--524",
journal = "International Journal of Pattern Recognition and Artificial Intelligence",
issn = "0218-0014",
publisher = "World Scientific Publishing",
number = "4",

}

User independent gesture interaction for small handheld devices. / Kallio, Sanna; Kela, Juha; Korpipää, Panu; Mäntyjärvi, Jani.

In: International Journal of Pattern Recognition and Artificial Intelligence, Vol. 20, No. 4, 2006, p. 505-524.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - User independent gesture interaction for small handheld devices

AU - Kallio, Sanna

AU - Kela, Juha

AU - Korpipää, Panu

AU - Mäntyjärvi, Jani

PY - 2006

Y1 - 2006

N2 - Accelerometer-based gesture recognition facilitates a complementary interaction modality for controlling mobile devices and home appliances. Using gestures for the task of home appliance control requires use of the same device and gestures by different persons, i.e. user independent gesture recognition. The practical application in small embedded low-resource devices also requires high computational performance. The user independent gesture recognition accuracy was evaluated with a set of eight gestures and seven users, with a total of 1120 gestures in the dataset. Twenty-state continuous HMM yielded an average of 96.9% user independent recognition accuracy, which was cross-validated by leaving one user in turn out of the training set. Continuous and discrete five-state HMM computational performances were compared with a reference test in a PC environment, indicating that discrete HMM is 20% faster. Computational performance of discrete five-state HMM was evaluated in an embedded hardware environment with a 104 MHz ARM-9 processor and Symbian OS. The average recognition time per gesture calculated from 1120 gesture repetitions was 8.3 ms. With this result, the computational performance difference between the compared methods is considered insignificant in terms of practical application. Continuous HMM is hence recommended as a preferred method due to its better suitability for a continuous-valued signal, and better recognition accuracy. The results suggest that, according to both evaluation criteria, HMM is feasible for practical user independent gesture control applications in mobile low-resource embedded environments.

AB - Accelerometer-based gesture recognition facilitates a complementary interaction modality for controlling mobile devices and home appliances. Using gestures for the task of home appliance control requires use of the same device and gestures by different persons, i.e. user independent gesture recognition. The practical application in small embedded low-resource devices also requires high computational performance. The user independent gesture recognition accuracy was evaluated with a set of eight gestures and seven users, with a total of 1120 gestures in the dataset. Twenty-state continuous HMM yielded an average of 96.9% user independent recognition accuracy, which was cross-validated by leaving one user in turn out of the training set. Continuous and discrete five-state HMM computational performances were compared with a reference test in a PC environment, indicating that discrete HMM is 20% faster. Computational performance of discrete five-state HMM was evaluated in an embedded hardware environment with a 104 MHz ARM-9 processor and Symbian OS. The average recognition time per gesture calculated from 1120 gesture repetitions was 8.3 ms. With this result, the computational performance difference between the compared methods is considered insignificant in terms of practical application. Continuous HMM is hence recommended as a preferred method due to its better suitability for a continuous-valued signal, and better recognition accuracy. The results suggest that, according to both evaluation criteria, HMM is feasible for practical user independent gesture control applications in mobile low-resource embedded environments.

KW - human computer interaction

KW - input technology

KW - mobile devices

KW - gesture recognition

KW - gesture control

KW - user interfaces

KW - user interaction technology

KW - pointing

KW - multimodal interaction

KW - accelerometer

U2 - 10.1142/S0218001406004776

DO - 10.1142/S0218001406004776

M3 - Article

VL - 20

SP - 505

EP - 524

JO - International Journal of Pattern Recognition and Artificial Intelligence

JF - International Journal of Pattern Recognition and Artificial Intelligence

SN - 0218-0014

IS - 4

ER -