Learning to detect video events from zero or very few video examples

作者:

Highlights:

• We deal with the challenging problem of high-level event detection in video.

• We build event detectors based solely on textual descriptions of the event classes.

• We also learn event detectors from very few positive and related training samples.

• We present results and comparisons on a large-scale TRECVID MED video dataset.

摘要

•We deal with the challenging problem of high-level event detection in video.•We build event detectors based solely on textual descriptions of the event classes.•We also learn event detectors from very few positive and related training samples.•We present results and comparisons on a large-scale TRECVID MED video dataset.

论文关键词:Video event detection,Textual event description,Zero positive examples,Few positive examples,Related videos

论文评审过程:Received 29 March 2015, Revised 9 September 2015, Accepted 28 September 2015, Available online 9 October 2015, Version of Record 9 September 2016.

论文官网地址:https://doi.org/10.1016/j.imavis.2015.09.005