CONF Varadarajan_BMVC2010_2010/IDIAP Probabilistic Latent Sequential Motifs: Discovering temporal activity patterns in video scenes Varadarajan, Jagannadan Emonet, Remi Odobez, Jean-Marc EXTERNAL https://publications.idiap.ch/attachments/papers/2010/Varadarajan_BMVC2010_2010.pdf PUBLIC https://publications.idiap.ch/index.php/publications/showcite/Varadarajan_Idiap-RR-33-2010 Related documents Aberystwyth University - BMVC 2010 Aberystwyth 2010 BMVA Press September 2010 117.1-117.11 This paper introduces a novel probabilistic activity modeling approach that mines recurrent sequential patterns from documents given as word-time occurrences. In this model, documents are represented as a mixture of sequential activity motifs (or topics) and their starting occurrences. The novelties are threefold. First, unlike previous approaches where topics only modeled the co-occurrence of words at a given time instant, our topics model the co-occurrence and temporal order in which the words occur within a temporal window. Second, our model accounts for the important case where activities occur concurrently in the document. And third, our method explicitly models with latent variables the starting time of the activities within the documents, enabling to implicitly align the occurrences of the same pattern during the joint inference of the temporal topics and their starting times. The model and its robustness to the presence of noise have been validated on synthetic data. Its effectiveness is also illustrated in video activity analysis from low-level motion features, where the discovered topics capture frequent patterns that implicitly represent typical trajectories of scene objects. REPORT Varadarajan_Idiap-RR-33-2010/IDIAP Probabilistic Latent Sequential Motifs: Discovering temporal activity patterns in video scenes Varadarajan, Jagannadan Emonet, Remi Odobez, Jean-Marc EXTERNAL https://publications.idiap.ch/attachments/reports/2010/Varadarajan_Idiap-RR-33-2010.pdf PUBLIC Idiap-RR-33-2010 2010 Idiap September 2010 This paper introduces a novel probabilistic activity modeling approach that mines recurrent sequential patterns from documents given as word-time occurrences. In this model, documents are represented as a mixture of sequential activity motifs (or topics) and their starting occurrences. The novelties are threefold. First, unlike previous approaches where topics only modeled the co-occurrence of words at a given time instant, our topics model the co-occurrence and temporal order in which the words occur within a temporal window. Second, our model counts for the important case where activities occur concurrently in the document. And third, our method explicitly models with latent variables the starting time of the activities within the documents, enabling to implicitly align the occurrences of the same pattern during the joint inference of the temporal topics and their starting times. The model and its robustness to the presence of noise have been validated on synthetic data. Its effectiveness is also illustrated in video activity analysis from low-level motion features, where the discovered topics capture frequent patterns that implicitly represent typical trajectories of scene objects.