Efficient Mining of Repetitions in Large-Scale TV Streams with Product Quantization Hashing

Jiangbo Yuan 1, 2 Guillaume Gravier 1 Sébastien Campion 1 Xiuwen Liu 2 Hervé Jégou 1
1 TEXMEX - Multimedia content-based indexing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : Duplicates or near-duplicates mining in video sequences is of broad interest to many multimedia applications. How to design an effective and scalable system, however, is still a challenge to the community. In this paper, we present a method to detect recurrent sequences in large-scale TV streams in an unsupervised manner and with little a priori knowledge on the content. The method relies on a product k-means quantizer that efficiently produces hash keys adapted to the data distribution for frame descriptors. This hashing technique combined with a temporal consistency check allows the detection of meaningful repetitions in TV streams. When considering all frames (about 47 millions) of a 22-day long TV broadcast, our system detects all repetitions in 15 minutes, excluding the computation of the frame descriptors. Experimental results show that our approach is a promising way to deal with very large video databases.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [18 references]  Display  Hide  Download


https://hal.inria.fr/hal-00731090
Contributor : Hervé Jégou <>
Submitted on : Wednesday, September 12, 2012 - 8:51:52 AM
Last modification on : Friday, November 16, 2018 - 1:21:57 AM
Document(s) archivé(s) le : Thursday, December 13, 2012 - 3:44:36 AM

Files

pqhash.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00731090, version 1

Citation

Jiangbo Yuan, Guillaume Gravier, Sébastien Campion, Xiuwen Liu, Hervé Jégou. Efficient Mining of Repetitions in Large-Scale TV Streams with Product Quantization Hashing. Workshop on Web-scale Vision and Social Media, in conjunction with ECCV, Oct 2012, Firenze, Italy. ⟨hal-00731090⟩

Share

Metrics

Record views

1046

Files downloads

447