Safely Caching HOG Pyramid Feature Levels, to Speed up Facial Landmark Detection
作者： Gareth Higgins , Ling Guan , Azhar Quddus , Ali Shahidi Zandi
DOI: 10.1007/978-3-030-27202-9_32
关键词:
摘要: A problem in object detection is finding the scale of interest, which often solved by multi-scale analysis. One drawbacks analysis that all scales are weighted equally important. We extend [7] to video data, using our method caches Histogram Oriented Graidents (HOG) feature levels. Utilizing a Bayesian Network (Bayes Net) discover policy for when partial or full pyramid updates required prevent loss tracking. Without sacrificing significant accuracy from original implementation.
springer.com
sci-hub.se 参考文章(6)
Lior Wolf, Tal Hassner, Itay Maoz, Face recognition in unconstrained videos with matched background similarity CVPR 2011. pp. 529- 534 ,(2011) , 10.1109/CVPR.2011.5995566
Xiangxin Zhu, D. Ramanan, Face detection, pose estimation, and landmark localization in the wild computer vision and pattern recognition. pp. 2879- 2886 ,(2012) , 10.1109/CVPR.2012.6248014
Tadas Baltrusaitis, Amir Zadeh, Yao Chong Lim, Louis-Philippe Morency, OpenFace 2.0: Facial Behavior Analysis Toolkit ieee international conference on automatic face gesture recognition. pp. 59- 66 ,(2018) , 10.1109/FG.2018.00019
Ali Farhadi, Joseph Redmon, YOLO9000: Better, Faster, Stronger arXiv: Computer Vision and Pattern Recognition. ,(2016)
Zuxuan Wu, Tushar Nagarajan, Abhishek Kumar, Steven Rennie, Larry S. Davis, Kristen Grauman, Rogerio Feris, BlockDrop: Dynamic Inference Paths in Residual Networks 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 8817- 8826 ,(2018) , 10.1109/CVPR.2018.00919
Joseph Redmon, Santosh Divvala, Ross Girshick, Ali Farhadi, You Only Look Once: Unified, Real-Time Object Detection computer vision and pattern recognition. pp. 779- 788 ,(2016) , 10.1109/CVPR.2016.91