Compact scalable hash from deep learning features aggregation for content de-duplication

Unprecedented growth in media content generation, communication and consumption has taken over the vast majority of storage spaces in devices, network caches, and clouds. How to identify duplications from network caches is an important issue for fast and efficient content delivery network (CDN) communication and storage. In this work, we developed a novel hash scheme which is scalable and robust to typical CDN induced transcoding and manipulations. Scalable hash design is constructed in essentially two stages: images are first represented as 512 channels of thumbnail images from the deep learning VGG-16 networks, and then a Fisher Vector aggregation is performed on the features which offer scalability in both underlying Gaussian Mixture Model (GMM) PCA embedding and component posterior likelihood. Hash is generated by direct binarizing the Fisher Vector with component/dimensionality priority optimization. Simulation results have demonstrated that this is a very compact and accurate scheme for CDN content de-duplication.

[1]  Refik Can Malli,et al.  Apparent Age Estimation Using Ensemble of Deep Learning Models , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[2]  Luis Enrique Sucar,et al.  Efficient video face recognition by using Fisher Vector encoding of binary features , 2016, 2016 23rd International Conference on Pattern Recognition (ICPR).

[3]  Aditya Akella,et al.  Redundancy in network traffic: findings and implications , 2009, SIGMETRICS '09.

[4]  Chu-Song Chen,et al.  Supervised Learning of Semantics-Preserving Hash via Deep Convolutional Neural Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  Meng Wang,et al.  Real-Time Video Copy-Location Detection in Large-Scale Repositories , 2011, IEEE MultiMedia.

[6]  Shogo Muramatsu,et al.  Interval calculation of EM algorithm for GMM parameter estimation , 2010, Proceedings of 2010 IEEE International Symposium on Circuits and Systems.

[7]  Luc Van Gool,et al.  Deep Features or Not: Temperature and Time Prediction in Outdoor Scenes , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[8]  Zhu Liu,et al.  Deep Hashing: A Joint Approach for Image Signature Learning , 2017, AAAI.

[9]  KyoungSoo Park,et al.  Effective content-based video caching with cache-friendly encoding and media-aware chunking , 2014, MMSys '14.

[10]  Jean-Luc Dugelay,et al.  Apparent Age Estimation from Face Images Combining General and Children-Specialized Deep Learning Models , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[11]  Florent Perronnin,et al.  Fisher Kernels on Visual Vocabularies for Image Categorization , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[12]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[13]  Srinivasan Seshan,et al.  Packet caches on routers: the implications of universal redundant traffic elimination , 2008, SIGCOMM '08.

[14]  Shanto Rahman,et al.  Application of deep learning to computer vision: A comprehensive study , 2016, 2016 5th International Conference on Informatics, Electronics and Vision (ICIEV).