Less Is More: Learning Highlight Detection From Video Duration

Bo Xiong, Yannis Kalantidis, Deepti Ghadiyaram, Kristen Grauman; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 1258-1267

Abstract


Highlight detection has the potential to significantly ease video browsing, but existing methods often suffer from expensive supervision requirements, where human viewers must manually identify highlights in training videos. We propose a scalable unsupervised solution that exploits video duration as an implicit supervision signal. Our key insight is that video segments from shorter user-generated videos are more likely to be highlights than those from longer videos, since users tend to be more selective about the content when capturing shorter videos. Leveraging this insight, we introduce a novel ranking framework that prefers segments from shorter videos, while properly accounting for the inherent noise in the (unlabeled) training data. We use it to train a highlight detector with 10M hashtagged Instagram videos. In experiments on two challenging public video highlight detection benchmarks, our method substantially improves the state-of-the-art for unsupervised highlight detection.

Related Material


[pdf]
[bibtex]
@InProceedings{Xiong_2019_CVPR,
author = {Xiong, Bo and Kalantidis, Yannis and Ghadiyaram, Deepti and Grauman, Kristen},
title = {Less Is More: Learning Highlight Detection From Video Duration},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
}