This paper presents a novel and practical face-assisted video coding scheme to enhance the visual quality of the face region in videophone applications. A low-complexity skin-color-based face detection and tracking scheme is proposed to locate the human face regions in realtime. After classifying the macroblocks (MBs) into the face and non-face regions, we present a dynamic distortion-weighting adjustment (DDWA) scheme to skip encoding the static non-face MBs, and the saved bits are used to compensate the face region by increasing the distortion weighting of the face MBs. The quality of the face regions will thus be largely enhanced. Moreover, the computation originally required for encoding the skipped MBs can also be saved. The experimental results show that the proposed method can significantly improve the PSNR and the subjective quality of face regions, while the degradation introduced on the non-face areas is relatively invisible to human perception. The proposed algorithm is fully compatible with the H. 263 standard, and the low complexity feature makes it well suited to be implemented for real-time applications.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Chia-Wen LIN, Yao-Jen CHANG, Yung-Chang CHEN, "A Low-Complexity Face-Assisted Coding Scheme for Low Bit-Rate Video Telephony" in IEICE TRANSACTIONS on Information,
vol. E86-D, no. 1, pp. 101-108, January 2003, doi: .
Abstract: This paper presents a novel and practical face-assisted video coding scheme to enhance the visual quality of the face region in videophone applications. A low-complexity skin-color-based face detection and tracking scheme is proposed to locate the human face regions in realtime. After classifying the macroblocks (MBs) into the face and non-face regions, we present a dynamic distortion-weighting adjustment (DDWA) scheme to skip encoding the static non-face MBs, and the saved bits are used to compensate the face region by increasing the distortion weighting of the face MBs. The quality of the face regions will thus be largely enhanced. Moreover, the computation originally required for encoding the skipped MBs can also be saved. The experimental results show that the proposed method can significantly improve the PSNR and the subjective quality of face regions, while the degradation introduced on the non-face areas is relatively invisible to human perception. The proposed algorithm is fully compatible with the H. 263 standard, and the low complexity feature makes it well suited to be implemented for real-time applications.
URL: https://globals.ieice.org/en_transactions/information/10.1587/e86-d_1_101/_p
Copy
@ARTICLE{e86-d_1_101,
author={Chia-Wen LIN, Yao-Jen CHANG, Yung-Chang CHEN, },
journal={IEICE TRANSACTIONS on Information},
title={A Low-Complexity Face-Assisted Coding Scheme for Low Bit-Rate Video Telephony},
year={2003},
volume={E86-D},
number={1},
pages={101-108},
abstract={This paper presents a novel and practical face-assisted video coding scheme to enhance the visual quality of the face region in videophone applications. A low-complexity skin-color-based face detection and tracking scheme is proposed to locate the human face regions in realtime. After classifying the macroblocks (MBs) into the face and non-face regions, we present a dynamic distortion-weighting adjustment (DDWA) scheme to skip encoding the static non-face MBs, and the saved bits are used to compensate the face region by increasing the distortion weighting of the face MBs. The quality of the face regions will thus be largely enhanced. Moreover, the computation originally required for encoding the skipped MBs can also be saved. The experimental results show that the proposed method can significantly improve the PSNR and the subjective quality of face regions, while the degradation introduced on the non-face areas is relatively invisible to human perception. The proposed algorithm is fully compatible with the H. 263 standard, and the low complexity feature makes it well suited to be implemented for real-time applications.},
keywords={},
doi={},
ISSN={},
month={January},}
Copy
TY - JOUR
TI - A Low-Complexity Face-Assisted Coding Scheme for Low Bit-Rate Video Telephony
T2 - IEICE TRANSACTIONS on Information
SP - 101
EP - 108
AU - Chia-Wen LIN
AU - Yao-Jen CHANG
AU - Yung-Chang CHEN
PY - 2003
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E86-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2003
AB - This paper presents a novel and practical face-assisted video coding scheme to enhance the visual quality of the face region in videophone applications. A low-complexity skin-color-based face detection and tracking scheme is proposed to locate the human face regions in realtime. After classifying the macroblocks (MBs) into the face and non-face regions, we present a dynamic distortion-weighting adjustment (DDWA) scheme to skip encoding the static non-face MBs, and the saved bits are used to compensate the face region by increasing the distortion weighting of the face MBs. The quality of the face regions will thus be largely enhanced. Moreover, the computation originally required for encoding the skipped MBs can also be saved. The experimental results show that the proposed method can significantly improve the PSNR and the subjective quality of face regions, while the degradation introduced on the non-face areas is relatively invisible to human perception. The proposed algorithm is fully compatible with the H. 263 standard, and the low complexity feature makes it well suited to be implemented for real-time applications.
ER -