Making FaceBoxes better and faster by reparameterization

Shuaihui Qi; Jintao Zhang; Zhiguang Bao; Dan Li

doi:10.1117/12.3011475

1 December 2023 Making FaceBoxes better and faster by reparameterization

Shuaihui Qi, Jintao Zhang, Zhiguang Bao, Dan Li

Proceedings Volume 12940, Third International Conference on Control and Intelligent Robotics (ICCIR 2023); 129402P (2023) https://doi.org/10.1117/12.3011475
Event: Third International Conference on Control and Intelligent Robotics (ICCIR 2023), 2023, Sipsongpanna, China

Abstract

Despite multi-branches network can strengthen the ability of feature extraction, the cost is speed reduction. Thus, in order to improve FaceBoxes’ accuracy and do not reduce speed, we propose Reparameterization-FaceBoxes in this paper that its multi-branches modules are replaced with diverse branch block. We make comparative experiments on WIDERFACE dataset, Reparameterization-FaceBoxes outperforms FaceBoxes with higher accuracy and faster speed, it can increase by 0.9%, 0.8% and 0.4% on Easy, Medium and Hard datasets while its speed increases about 20% for VGA-resolution images.

(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Shuaihui Qi, Jintao Zhang, Zhiguang Bao, and Dan Li "Making FaceBoxes better and faster by reparameterization", Proc. SPIE 12940, Third International Conference on Control and Intelligent Robotics (ICCIR 2023), 129402P (1 December 2023); https://doi.org/10.1117/12.3011475

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available