Deep Reinforcement Learning for Unsupervised Video Summarization with Diversity-Representativeness Reward

By Kaiyang Zhou, Yu Qiao. AAAI Conference on Artificial Intelligence, 2017

509 citations recorded by the source corpus.

Video summarization aims to facilitate large-scale video browsing by producing short, concise summaries that are diverse and representative of original videos. In this paper, we formulate video summarization as a sequential decision-making process and develop a deep summarization network (DSN) to summarize videos. DSN predicts for each video frame a probability, which indicates how likely a frame is selected, and then takes actions based on the probability distributions to select frames, forming video summaries. To train our DSN, we propose an end-to-end, reinforcement learning-based framework, where we design a novel reward function that jointly accounts for diversity and representativeness of generated summaries and does not rely on labels or user interactions at all. During training, the reward function judges how diverse and representative the generated summaries are, while DSN strives for earning higher rewards by learning to produce more diverse and more representative summaries. Since labels are not required, our method can be fully unsupervised. Extensive experiments on two benchmark datasets show that our unsupervised method not only outperforms other state-of-the-art unsupervised methods, but also is comparable to or even superior than most of published supervised approaches.

Subfields: Computer Vision and Pattern Recognition, Signal Processing

View Paper: https://www.semanticscholar.org/paper/9e9e5f0c36548cfe2855aae46b519b146aa8c9ae

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

Video Summarization Using Deep Neural Networks: A Survey (2021, methodology): “splits; 5-fold cross validation), the way that the computed FScores from the pairwise comparisons with the different user summaries are taken under consideration (maximum value is kept for SumMe according to [100]; average value is kept for TVSum) to form the F-Score for a given test video, and the ”
Deep Attentive Video Summarization With Distribution Consistency Learning (2020, methodology): “loss, especially mse, is the most popular loss employed in video summarization [2], [3], [10], [29], and it minimizes the”