Real-time Detection of Activities in Untrimmed Videos
2020 IEEE Winter Applications of Computer Vision Workshops (WACVW)(2020)
摘要
Real-time detection of spatio-temporal sparse activities in untrimmed videos is a challenging problem. In this work, we present the details of our proposed solution. We begin with a slow baseline implementation of a previously state-of-the-art system [13] and redesign it to achieve real-time performance for detecting 37 activities in the ActEV19 Sequestered Data Leaderboard [4]. This is primarily achieved by introducing speed related hyperparameters into the baseline approach. A tradeoff analysis is performed to assist in hyperparameter selection which results in a real-time, high quality action detection system. Our system achieves an AUDC score of 0.476 on the ActEV19 Sequestered Data Leaderboard.
更多查看译文
关键词
untrimmed videos,real-time detection,spatio-temporal sparse activities,slow baseline implementation,state-of-the-art system,real-time performance,speed related hyperparameters,baseline approach,hyperparameter selection,high quality action detection system,ActEV19 Sequestered Data Leaderboard
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络