Loading...

A new two-stage topic model based framework for modeling traffic motion patterns

Ahmadi, P ; Sharif University of technology | 2018

795 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/IranianMVIP.2017.8342356
  3. Publisher: IEEE Computer Society , 2018
  4. Abstract:
  5. Analyzing motion patterns in traffic videos can be exploited directly to generate high-level descriptions of the video content. The most recent and successful unsupervised methods for complex traffic scene analysis are based on topic models. In this paper, a new two-stage framework is proposed for traffic motion pattern extraction based on topic models. This framework forces the topic model to learn known meaningful motion patterns in traffic scenes. Latent Dirichlet Allocation (LDA) is employed as the topic model. Experimental results show that our proposed framework finds the motion patterns more efficiently and gives a meaningful representation for the video. © 2017 IEEE
  6. Keywords:
  7. Latent Dirichlet Allocation (LDA) ; Motion patterns ; Topic model ; Computer vision ; Statistics ; Time and motion study ; High level description ; Latent dirichlet allocations ; Motion pattern ; Topic Modeling ; Traffic scene ; Traffic scene analysis ; Traffic videos ; Unsupervised method ; Data mining
  8. Source: 10th Iranian Conference on Machine Vision and Image Processing, MVIP 2017 ; Volume 2017-November , April , 2018 , Pages 276-280 ; 21666776 (ISSN) ; 9781538644041 (ISBN)
  9. URL: https://ieeexplore.ieee.org/document/8342356