光电工程, 2020, 47 (1): 190135, 网络出版: 2020-02-24  

级联网络和金字塔光流的旋转不变人脸检测

Rotating invariant face detection via cascaded networks and pyramidal optical flows
孙锐 1,2,*阚俊松 1,2吴柳玮 1,2王鹏 3
作者单位
1 合肥工业大学计算机与信息学院, 安徽合肥 230009
2 工业安全与应急技术安徽省重点实验室, 安徽合肥 230009
3 合肥进毅智能技术有限公司, 安徽合肥 230088
摘要
在无约束的开放空间中, 由于面部姿态变化、背景环境复杂、运动模糊等, 人脸检测仍是一个具有挑战性的任务。本文针对视频流中人脸检测存在的平面内旋转问题, 将人脸关键点与金字塔光流相结合, 提出了基于级联网络和金字塔光流的旋转不变人脸检测算法。首先利用级联渐进卷积神经网络对视频流中前一帧进行人脸位置和关键点的定位; 其次为获取关键点与人脸候选框间光流映射, 使用独立的关键点检测网络对当前帧进行再次定位; 之后计算前后两帧之间关键点光流位移; 最后通过关键点光流位移与人脸候选框的映射关系, 对视频中检测到的人脸进行校正, 从而完成平面内旋转人脸不变性检测。实验经 FDDB公开数据集上测试, 证明该方法精确度较高。并且, 在 Boston面部跟踪数据集上进行动态测试, 证明该人脸检测算法能有效解决平面内旋转人脸检测问题。对比其它检测算法, 该算法检测速度有较大优势, 同时视频中窗口抖动问题得到了很好解决。
Abstract
In the unconstrained open-space, face detection is still a challenging task due to the facial posture changes, complex background environment, and motion blur. The rotation-invariant algorithm based on cascaded network and pyramid optical flow is proposed. Firstly, the cascading progressive convolutional neural network is adopted to locate the face position and facial landmark of the previous frame in the video stream. Secondly, the in-dependent facial landmark detection network is used to reposition the current frame, and the optical flow mapping displacement of the facial landmark between the two frames is calculated afterwards. Finally, the detected face is corrected by the mapping relationship between the optical flow displacement of the facial landmark and the bounding box, thereby completing the rotation-invariant face detection. The experiment was tested on the FDDB public data-sets, which proved that the method is more accurate. Moreover, the dynamic test on the Boston head tracking da-tabase proves that the face detection algorithm can effectively solve the problem of rotation-invariant face detection. Compared with other detection algorithms, the detection speed of the proposed algorithm has a great advantage, and the window jitter problem in the video is well solved.

孙锐, 阚俊松, 吴柳玮, 王鹏. 级联网络和金字塔光流的旋转不变人脸检测[J]. 光电工程, 2020, 47(1): 190135. Sun Rui, Kan Junsong, Wu Liuwei, Wang Peng. Rotating invariant face detection via cascaded networks and pyramidal optical flows[J]. Opto-Electronic Engineering, 2020, 47(1): 190135.

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!