Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks

Abstract: 

Predicting the gaze of a user can have important applications in hu- man computer interactions (HCI). They find applications in areas such as social interaction, driver distraction, human robot interaction and education. Appearance based models for gaze estimation have significantly improved due to recent advances in convolutional neural network (CNN). This paper proposes a method to predict the gaze of a user with deep models purely based on CNNs. A key novelty of the proposed model is that it produces a probabilistic map describing the gaze distribution (as opposed to predicting a single gaze direction). This approach is achieved by converting the regres- sion problem into a classification problem, predicting the probabil- ity at the output instead of a single direction. The framework relies in a sequence of downsampling followed by upsampling to obtain the probabilistic gaze map. We observe that our proposed approach works better than a regression model in terms of prediction accuracy. The average mean squared error between the predicted gaze and the true gaze is observed to be 6.89◦ in a model trained and tested on the MSP-Gaze database, without any calibration or adaptation to the target user.

up
0 users have voted:

Paper Details

Authors:
Sumit Jha, Carlos Busso
Submitted On:
20 May 2020 - 9:53am
Short Link:
Type:
Poster
Event:
Presenter's Name:
Sumit Jha
Document Year:
2019
Cite

Document Files

Jha_2019-poster.pdf

(34)

Subscribe

[1] Sumit Jha, Carlos Busso, "Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5410. Accessed: Aug. 14, 2020.
@article{5410-20,
url = {http://sigport.org/5410},
author = {Sumit Jha; Carlos Busso },
publisher = {IEEE SigPort},
title = {Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks},
year = {2020} }
TY - EJOUR
T1 - Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks
AU - Sumit Jha; Carlos Busso
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5410
ER -
Sumit Jha, Carlos Busso. (2020). Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks. IEEE SigPort. http://sigport.org/5410
Sumit Jha, Carlos Busso, 2020. Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks. Available at: http://sigport.org/5410.
Sumit Jha, Carlos Busso. (2020). "Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks." Web.
1. Sumit Jha, Carlos Busso. Estimation of gaze region using two dimensional probabilistic maps constructed using convolutional neural networks [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5410