Event Detection in Continuous Video: An Inference in Point Process Approach (2017)

by Zhen Qin and Christian R. Shelton


Abstract: We propose a novel approach towards event detection in real-world continuous video sequences. The method 1) is able to model arbitrary-order non-Markovian dependencies in videos to mitigate local visual ambiguities, 2) conducts simultaneous event segmentation and labeling, and 3) is time-window free. The idea is to represent a video as an event stream of both high-level semantic events and low-level video observations. In training, we learn a point process model called piecewise-constant conditional intensity model (PCIM) that is able to capture complex non-Markovian dependencies in the event streams. In testing, event detection can be modeled as the inference of high-level semantic events, given low-level image observations. We develop the first inference algorithm for PCIM and show it samples exactly from the posterior distribution. We then evaluate the video event detection task on real-world video sequences. Our model not only provides competitive results on the video event segmentation and labeling task, but also provides benefits including being interpretable and efficient.


Download Information

Zhen Qin and Christian R. Shelton (2017). "Event Detection in Continuous Video: An Inference in Point Process Approach." IEEE Transactions on Image Processing, 26(12), 5680-5691. pdf          

Bibtex citation

@article{QinShe17,
   author = "Zhen Qin and Christian R. Shelton",
   title = "Event Detection in Continuous Video: An Inference in Point Process Approach",
   journal = "IEEE Transactions on Image Processing",
   journalabbr = "TIP",
   month = dec,
   year = 2017,
   volume = 26,
   number = 12,
   pages = "5680--5691",
}