publication . Conference object . 2016

Deep Multi-task Learning with Label Correlation Constraint for Video Concept Detection

Foteini Markatopoulou; Vasileios Mezaris; Ioannis Patras;
Open Access
  • Published: 17 Oct 2016
Abstract
In this work we propose a method that integrates multi-task learning (MTL) and deep learning. Our method appends a MTL-like loss to a deep convolutional neural network, in order to learn the relations between tasks together at the same time, and also incorporates the label correlations between pairs of tasks. We apply the proposed method on a transfer learning scenario, where our objective is to fine-tune the parameters of a network that has been originally trained on a large-scale image dataset for concept detection, so that it be applied on a target video dataset and a corresponding new set of target concepts. We evaluate the proposed method for the video conc...
Persistent Identifiers
Subjects
free text keywords: Concept detection; deep learning; video analysis, Deep learning, Transfer of learning, Pattern recognition, Computer science, Artificial intelligence, business.industry, business, TRECVID, Convolutional neural network, Search engine indexing, Machine learning, computer.software_genre, computer, Multi-task learning, Correlation, Video annotation
Funded by
EC| InVID
Project
InVID
In Video Veritas – Verification of Social Media Video Content for the News Industry
  • Funder: European Commission (EC)
  • Project Code: 687786
  • Funding stream: H2020 | IA
Validated by funder
,
EC| MOVING
Project
MOVING
Training towards a society of data-savvy information professionals to enable open leadership innovation
  • Funder: European Commission (EC)
  • Project Code: 693092
  • Funding stream: H2020 | RIA
Validated by funder
Download fromView all 4 versions
Open Access
ZENODO
Conference object . 2016
Provider: ZENODO
Restricted
https://dl.acm.org/doi/pdf/10....
Conference object . 2016
Provider: Crossref
Any information missing or wrong?Report an Issue