Triplet Loss Feature Aggregation For Scalable Hash

The increasing demands of high resolution and quality aggravate the status of heavy burden of cluster storage side and restricted bandwidth resources. Hence, video de-duplication in storage and transmission is becoming an important feature for video cloud storage and Content Delivery Network (CDN) service providers. The current video de-duplication schemes mostly relies on the URL based solution, which is not able to deal with non-cacheable content like video, which the same piece of content may have totally different URL identification and fragmentation and different quality representations further complicate the problem. In this paper, we propose a novel content based video segmentation identification scheme that is invariant to the underlying codec and operational bit rates, it computes robust features from a triplet loss deep learning network that captures the invariance of the same content under different coding tools and strategy, while a scalable hashing solution is developed based on Fisher Vector aggregation of the convolutional features from the Triplet loss network. Moreover, we apply binary tree to obtain the triplets to improve the performance of the triplet-loss based VGG network. Our simulation results demonstrate the great improvement in terms of large scale video repository de-duplication compared with state-of-the-art methods.