publication . Conference object . 2019

A Stepwise, Label-based Approach for Improving the Adversarial Training in Unsupervised Video Summarization

Evlampios E. Apostolidis; Ioannis Patras; Eleni Adamantidou; Vasileios Mezaris; Alexandros I. Metsai;
Open Access
  • Published: 21 Oct 2019
Abstract
In this paper we present our work on improving the efficiency of adversarial training for unsupervised video summarization. Our starting point is the SUM-GAN model, which creates a representative summary based on the intuition that such a summary should make it possible to reconstruct a video that is indistinguishable from the original one. We build on a publicly available implementation of a variation of this model, that includes a linear compression layer to reduce the number of learned parameters and applies an incremental approach for training the different components of the architecture. After assessing the impact of these changes to the model's performance...
Persistent Identifiers
Subjects
free text keywords: Video Summarization, Unsupervised Learning, Adversarial Training, Evaluation Protocol, Datasets, Artificial intelligence, business.industry, business, Intuition, Architecture, Adversarial system, Automatic summarization, Benchmarking, Machine learning, computer.software_genre, computer, Computer science
Funded by
EC| ReTV
Project
ReTV
Enhancing and Re-Purposing TV Content for Trans-Vector Engagement
  • Funder: European Commission (EC)
  • Project Code: 780656
  • Funding stream: H2020 | RIA
Download fromView all 2 versions
ZENODO
Conference object . 2019
Provider: ZENODO
http://dl.acm.org/ft_gateway.c...
Conference object . 2019
Provider: Crossref
Any information missing or wrong?Report an Issue