Temporal Action Localization by Structured Maximal Sums
This addresses the problem of accurately detecting and localizing actions in videos for applications like video analysis and surveillance, representing an incremental improvement with a novel structured approach.
The paper tackles temporal action localization in videos by modeling actions as structured predictions over temporal windows and developing an efficient algorithm for finding structured maximal sums, achieving competitive performance on the THUMOS 14 benchmark.
We address the problem of temporal action localization in videos. We pose action localization as a structured prediction over arbitrary-length temporal windows, where each window is scored as the sum of frame-wise classification scores. Additionally, our model classifies the start, middle, and end of each action as separate components, allowing our system to explicitly model each action's temporal evolution and take advantage of informative temporal dependencies present in this structure. In this framework, we localize actions by searching for the structured maximal sum, a problem for which we develop a novel, provably-efficient algorithmic solution. The frame-wise classification scores are computed using features from a deep Convolutional Neural Network (CNN), which are trained end-to-end to directly optimize for a novel structured objective. We evaluate our system on the THUMOS 14 action detection benchmark and achieve competitive performance.