A Panoramic Video Face Detection System Design and Implement

A panorama is a wide-angle view picture with high-resolution, usually composed of multiple images, and has a wide range of applications in surveillance and entertainment. This paper presents a end-to-end real-time panoramic face detection video system, which generates panorama video efficiently and effectively with the ability of face detection. We fix the relative position of the camera and use the speeded up robust features (SURF) matching algorithm to calibrate the cameras in the offline stage. In the online stage, we improve the parallel execution speed of image stitching using the latest compute unified device architecture (CUDA) technology. The proposed design fulfils high-quality automatic image stitching algorithm to provide a seamless panoramic image with 6k resolution at 25 fps. We also design a convolutional neural network to build a face detection model suitable for panorama input. The model performs very well especially in small faces and multi-faces, and can maintain the detection speed of 25 fps at high resolution.

[1]  Matthew A. Brown,et al.  Automatic Panoramic Image Stitching using Invariant Features , 2007, International Journal of Computer Vision.

[2]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  Pritam Prakash Shete,et al.  Real-time panorama composition for video surveillance using GPU , 2016, 2016 International Conference on Advances in Computing, Communications and Informatics (ICACCI).

[4]  Tomaso A. Poggio,et al.  A general framework for object detection , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[5]  Yoav Freund,et al.  A decision-theoretic generalization of on-line learning and an application to boosting , 1997, EuroCOLT.

[6]  Fernando De la Torre,et al.  Supervised Descent Method and Its Applications to Face Alignment , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[7]  Yu Qiao,et al.  Joint Face Detection and Alignment Using Multitask Cascaded Convolutional Networks , 2016, IEEE Signal Processing Letters.

[8]  Ali Farhadi,et al.  You Only Look Once: Unified, Real-Time Object Detection , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[9]  Paul A. Viola,et al.  Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[10]  Luc Van Gool,et al.  The Pascal Visual Object Classes (VOC) Challenge , 2010, International Journal of Computer Vision.

[11]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[12]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[13]  Bruno Feijó,et al.  Real time 360° video stitching and streaming , 2016, SIGGRAPH Posters.

[14]  Younghui Kim,et al.  Rich360 , 2016, ACM Trans. Graph..

[15]  Wei Liu,et al.  SSD: Single Shot MultiBox Detector , 2015, ECCV.