3D head pose estimation with convolutional neural network trained on synthetic images

Xiabing Liu; Wei Liang; Yumeng Wang; Shuyang Li; Mingtao Pei

doi:10.1109/ICIP.2016.7532566

3D head pose estimation with convolutional neural network trained on synthetic images

Xiabing Liu
, Wei Liang
, Yumeng Wang
, Shuyang Li
, Mingtao Pei

School of Computer Science and Technology

Beijing Institute of Technology

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

88 Citations (Scopus)

Abstract

In this paper, we propose a method to estimate head pose with convolutional neural network, which is trained on synthetic head images. We formulate head pose estimation as a regression problem. A convolutional neural network is trained to learn head features and solve the regression problem. To provide annotated head poses in the training process, we generate a realistic head pose dataset by rendering techniques, in which we consider the variation of gender, age, race and expression. Our dataset includes 74000 head poses rendered from 37 head models. For each head pose, RGB image and annotated pose parameters are given. We evaluate our method on both synthetic and real data. The experiments show that our method improves the accuracy of head pose estimation.

Original language	English
Title of host publication	2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings
Publisher	IEEE Computer Society
Pages	1289-1293
Number of pages	5
ISBN (Electronic)	9781467399616
DOIs	https://doi.org/10.1109/ICIP.2016.7532566
Publication status	Published - 3 Aug 2016
Event	23rd IEEE International Conference on Image Processing, ICIP 2016 - Phoenix, United States Duration: 25 Sept 2016 → 28 Sept 2016

Publication series

Name	Proceedings - International Conference on Image Processing, ICIP
Volume	2016-August
ISSN (Print)	1522-4880

Conference

Conference	23rd IEEE International Conference on Image Processing, ICIP 2016
Country/Territory	United States
City	Phoenix
Period	25/09/16 → 28/09/16

Keywords

Convolutional neural network
Head pose estimation
Synthetic images

Access to Document

10.1109/ICIP.2016.7532566

Cite this

Liu, X., Liang, W., Wang, Y., Li, S., & Pei, M. (2016). 3D head pose estimation with convolutional neural network trained on synthetic images. In 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings (pp. 1289-1293). Article 7532566 (Proceedings - International Conference on Image Processing, ICIP; Vol. 2016-August). IEEE Computer Society. https://doi.org/10.1109/ICIP.2016.7532566

@inproceedings{e43e040340ec4e549e7e2c0969b280c5,

title = "3D head pose estimation with convolutional neural network trained on synthetic images",

abstract = "In this paper, we propose a method to estimate head pose with convolutional neural network, which is trained on synthetic head images. We formulate head pose estimation as a regression problem. A convolutional neural network is trained to learn head features and solve the regression problem. To provide annotated head poses in the training process, we generate a realistic head pose dataset by rendering techniques, in which we consider the variation of gender, age, race and expression. Our dataset includes 74000 head poses rendered from 37 head models. For each head pose, RGB image and annotated pose parameters are given. We evaluate our method on both synthetic and real data. The experiments show that our method improves the accuracy of head pose estimation.",

keywords = "Convolutional neural network, Head pose estimation, Synthetic images",

author = "Xiabing Liu and Wei Liang and Yumeng Wang and Shuyang Li and Mingtao Pei",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 23rd IEEE International Conference on Image Processing, ICIP 2016 ; Conference date: 25-09-2016 Through 28-09-2016",

year = "2016",

month = aug,

day = "3",

doi = "10.1109/ICIP.2016.7532566",

language = "English",

series = "Proceedings - International Conference on Image Processing, ICIP",

publisher = "IEEE Computer Society",

pages = "1289--1293",

booktitle = "2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings",

address = "United States",

}

Liu, X, Liang, W, Wang, Y, Li, S & Pei, M 2016, 3D head pose estimation with convolutional neural network trained on synthetic images. in 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings., 7532566, Proceedings - International Conference on Image Processing, ICIP, vol. 2016-August, IEEE Computer Society, pp. 1289-1293, 23rd IEEE International Conference on Image Processing, ICIP 2016, Phoenix, United States, 25/09/16. https://doi.org/10.1109/ICIP.2016.7532566

3D head pose estimation with convolutional neural network trained on synthetic images. / Liu, Xiabing; Liang, Wei; Wang, Yumeng et al.
2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. IEEE Computer Society, 2016. p. 1289-1293 7532566 (Proceedings - International Conference on Image Processing, ICIP; Vol. 2016-August).

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

TY - GEN

T1 - 3D head pose estimation with convolutional neural network trained on synthetic images

AU - Liu, Xiabing

AU - Liang, Wei

AU - Wang, Yumeng

AU - Li, Shuyang

AU - Pei, Mingtao

PY - 2016/8/3

Y1 - 2016/8/3

N2 - In this paper, we propose a method to estimate head pose with convolutional neural network, which is trained on synthetic head images. We formulate head pose estimation as a regression problem. A convolutional neural network is trained to learn head features and solve the regression problem. To provide annotated head poses in the training process, we generate a realistic head pose dataset by rendering techniques, in which we consider the variation of gender, age, race and expression. Our dataset includes 74000 head poses rendered from 37 head models. For each head pose, RGB image and annotated pose parameters are given. We evaluate our method on both synthetic and real data. The experiments show that our method improves the accuracy of head pose estimation.

AB - In this paper, we propose a method to estimate head pose with convolutional neural network, which is trained on synthetic head images. We formulate head pose estimation as a regression problem. A convolutional neural network is trained to learn head features and solve the regression problem. To provide annotated head poses in the training process, we generate a realistic head pose dataset by rendering techniques, in which we consider the variation of gender, age, race and expression. Our dataset includes 74000 head poses rendered from 37 head models. For each head pose, RGB image and annotated pose parameters are given. We evaluate our method on both synthetic and real data. The experiments show that our method improves the accuracy of head pose estimation.

KW - Convolutional neural network

KW - Head pose estimation

KW - Synthetic images

UR - https://www.scopus.com/pages/publications/85006736322

U2 - 10.1109/ICIP.2016.7532566

DO - 10.1109/ICIP.2016.7532566

M3 - Conference contribution

AN - SCOPUS:85006736322

T3 - Proceedings - International Conference on Image Processing, ICIP

SP - 1289

EP - 1293

BT - 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings

PB - IEEE Computer Society

T2 - 23rd IEEE International Conference on Image Processing, ICIP 2016

Y2 - 25 September 2016 through 28 September 2016

ER -

3D head pose estimation with convolutional neural network trained on synthetic images

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this