Abstract | ||
---|---|---|
In the current context of monetization of multimedia content, it is common to see the appearance of edited replicas of popular videos to take advantage of the momentum of those. In this work, several parameters of near-duplicate video detection systems based on codebooks are studied using techniques from the field of information retrieval. As a result, a system with high average precision, usually higher than 85%, is obtained. Several hyperparameters of the system, such as the aggregation mechanisms and the retrieval model, are analyzed, thus adjusting the system for optimal performance. |
Year | DOI | Venue |
---|---|---|
2021 | 10.1007/978-3-030-87869-6_27 | 16TH INTERNATIONAL CONFERENCE ON SOFT COMPUTING MODELS IN INDUSTRIAL AND ENVIRONMENTAL APPLICATIONS (SOCO 2021) |
Keywords | DocType | Volume |
Near-duplicates, Video analysis, Convolutional neural network, Transfer learning | Conference | 1401 |
ISSN | Citations | PageRank |
2194-5357 | 0 | 0.34 |
References | Authors | |
0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Guillermo Hernández | 1 | 3 | 2.06 |
Angélica González Arrieta | 2 | 31 | 10.04 |
Paulo Novais | 3 | 0 | 1.01 |
Sara Rodríguez | 4 | 0 | 0.34 |