Video Summarisation by Classification with Deep Reinforcement Learning
This addresses the problem of creating informative video summaries for users in multimedia applications, but it is incremental as it builds on existing weakly supervised approaches.
The paper tackled video summarization by proposing a reinforcement learning method that uses video-level category labels to generate summaries with category-related information, achieving state-of-the-art performance on two benchmark datasets.
Most existing video summarisation methods are based on either supervised or unsupervised learning. In this paper, we propose a reinforcement learning-based weakly supervised method that exploits easy-to-obtain, video-level category labels and encourages summaries to contain category-related information and maintain category recognisability. Specifically, We formulate video summarisation as a sequential decision-making process and train a summarisation network with deep Q-learning (DQSN). A companion classification network is also trained to provide rewards for training the DQSN. With the classification network, we develop a global recognisability reward based on the classification result. Critically, a novel dense ranking-based reward is also proposed in order to cope with the temporally delayed and sparse reward problems for long sequence reinforcement learning. Extensive experiments on two benchmark datasets show that the proposed approach achieves state-of-the-art performance.