Predicting Perceptual Haptic Attributes of Textured Surface from Tactile Data Based on Deep CNN-LSTM Network
Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Standard
Predicting Perceptual Haptic Attributes of Textured Surface from Tactile Data Based on Deep CNN-LSTM Network. / Awan, Mudassir Ibrahim; Hassan, Waseem; Jeon, Seokhee.
VRST 2023 - 29th ACM Symposium on Virtual Reality Software and Technology. Association for Computing Machinery, Inc., 2023. 33 (Proceedings of the ACM Symposium on Virtual Reality Software and Technology, VRST).Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Predicting Perceptual Haptic Attributes of Textured Surface from Tactile Data Based on Deep CNN-LSTM Network
AU - Awan, Mudassir Ibrahim
AU - Hassan, Waseem
AU - Jeon, Seokhee
N1 - Funding Information: This research was supported by the Ministry of Science and ICT Korea under the ITRC support program (IITP-2023-RS-2022-00156354), under the IITP program (2022-0-01005), both supervised by IITP, and under the Mid-Researcher Program (2022R1A2C1008483) supervised by the NRF Korea. Publisher Copyright: © 2023 ACM.
PY - 2023/10/9
Y1 - 2023/10/9
N2 - This paper introduces a framework to predict multi-dimensional haptic attribute values that humans use to recognize the material by using the physical tactile signals (acceleration) generated when a textured surface is stroked. To this end, two spaces are established: a haptic attribute space and a physical signal space. A five-dimensional haptic attribute space is established through human adjective rating experiments with the 25 real texture samples. The physical space is constructed using tool-based interaction data from the same 25 samples. A mapping is modeled between the aforementioned spaces using a newly designed CNN-LSTM deep learning network. Finally, a prediction algorithm is implemented that takes acceleration data and returns coordinates in the haptic attribute space. A quantitative evaluation was conducted to inspect the reliability of the algorithm on unseen textures, showing that the model outperformed other similar models.
AB - This paper introduces a framework to predict multi-dimensional haptic attribute values that humans use to recognize the material by using the physical tactile signals (acceleration) generated when a textured surface is stroked. To this end, two spaces are established: a haptic attribute space and a physical signal space. A five-dimensional haptic attribute space is established through human adjective rating experiments with the 25 real texture samples. The physical space is constructed using tool-based interaction data from the same 25 samples. A mapping is modeled between the aforementioned spaces using a newly designed CNN-LSTM deep learning network. Finally, a prediction algorithm is implemented that takes acceleration data and returns coordinates in the haptic attribute space. A quantitative evaluation was conducted to inspect the reliability of the algorithm on unseen textures, showing that the model outperformed other similar models.
KW - Haptic texture classification
KW - neural network
KW - psychophysics
U2 - 10.1145/3611659.3615714
DO - 10.1145/3611659.3615714
M3 - Article in proceedings
AN - SCOPUS:85175260496
T3 - Proceedings of the ACM Symposium on Virtual Reality Software and Technology, VRST
BT - VRST 2023 - 29th ACM Symposium on Virtual Reality Software and Technology
PB - Association for Computing Machinery, Inc.
T2 - 29th ACM Symposium on Virtual Reality Software and Technology, VRST 2023
Y2 - 9 October 2023 through 11 October 2023
ER -
ID: 388954624