5/7/2023 0 Comments Duplicacy hash option![]() ![]() ![]() ![]() Especially, we focus on the true positive rate (TPR) under false positive rate equals 0 because we cannot allow the judgment is error and the product is deleted accidentally. This results in that we should derive the system of high accuracy and recall. ![]() First of all, multimedia data on the cloud cluster and CDN is all the cherish product from industry and user, so it is extremely strict to remove any videos. In spite of that leveraging the video de-duplication scheme is quite necessary and promising, the micro improvement of its performance exhibits it is difficult to develop. Hence, how to retrieve and remove the duplicated versions of videos is an essential task for researchers. If users apply the same version (REQP) of videos from the server ignoring the identical ones in the content delivery network (CDN), the pressure on the network from video delivery and storage will be quite large. We define the version as the combination of resolution and quantization parameter namely REQP in this paper. In current media content storage scheme, the storage side has to hold all of the REQP media content, which is error-prone and not cost-effective. Fig. 1 depicts that various contents of resolutions and quantized parameters(REQP) are consumed by very diversified consumers’ platform. Therefore the massive multimedia data is pushing forward the paradigm of effective storage on cluster servers. Scalable and robust signatures for media content to support de-duplication at fine granular spatio-temporal segments granularity, are important to rip the full benefits of storage de-duplication. New compact rate agnostic and coding scheme agnostic content identification and hashing solution are needed, to characterize media segments across different representations and with totally different bit streams. This creates challenges to the existing Content Delivery Network (CDN) and storage de-duplication schemes like those based on MD5 hashing of file chunks. If a content identification scheme can support identification of duplicates in network caches in core networks and edge nodes, then traffic can be localized and bandwidth saved. There is a de-duplication of media content use case for example. The media content creation, sharing and consumption process generate many duplicates but are not necessarily identical in bit stream. OTT (over the top) content providers are also pushing subscription-based video on demand (VoD) services that offer streaming services on television. Modern dynamic adaptive video streaming methods such as MPEG-DASH , Apple HLS and Microsoft Smooth Streaming have a great impact on how content providers store and serve the media contents in the cloud, such as a content delivery network (CDN). Our simulation results demonstrate the great improvement in terms of large scale video repository de-duplication compared with state-of-the-art methods. 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. 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. Despite of the necessity of optimizing the multimedia data de-duplication approach, it is a challenging task because we should match as many as possible duplicated videos under not removing videos by mistake. Hence, video de-duplication in storage and transmission is becoming an important feature for video cloud storage and Content Delivery Network (CDN) service providers. Moreover, the increasing demands of high resolution and quality aggravate the status of heavy burden of cluster storage side and restricted bandwidth resources. The inefficiency of this duplicates to storage and communication motivate researchers in both academia and industry to come up with computationally efficient video deduplication solutions for storage and CDN providers. The producing, sharing and consuming life cycle of video content creates massive amount of duplicates in video segments due to variable bit rate representation and fragmentation in the playbacks. ![]()
0 Comments
Leave a Reply. |