Results 1 -
5 of
5
Efficient activity retrieval through semantic graph queries
- In ACM Multimedia
, 2015
"... We present an efficient retrieval approach for activity detection in large surveillance video datasets based on semantic graph queries. Unlike conventional approaches, our zero-shot retrieval method does not require knowledge of the activity classes contained in the video. We propose a novel user-ce ..."
Abstract
-
Cited by 2 (2 self)
- Add to MetaCart
(Show Context)
We present an efficient retrieval approach for activity detection in large surveillance video datasets based on semantic graph queries. Unlike conventional approaches, our zero-shot retrieval method does not require knowledge of the activity classes contained in the video. We propose a novel user-centric approach that models queries through the creation of sparse semantic graphs based on attributes and dis-criminative relationships. We then pose search as a ranked sub-graph matching problem and leverage the fact that the attributes and relationships in the query have different levels of discriminabil-ity to filter out bad matches. Rather than solving the NP-hard ex-act subgraph matching problem, we develop a novel maximally discriminative spanning tree (MDST) as the relaxation of a given query graph, and then describe a matching algorithm that recovers matches to this tree in linear time using maximally discriminative subgraph matching (MDSM). We utilize the MDST to minimize the number of possible matches to the original query while guarantee-ing that the best matches are within this set. We test this algorithm on two large video datasets: the 35-GB Virat Ground dataset and a 1-TB aerial data collection from Yuma. These datasets yield graphs with 200,000 nodes and 1 million nodes, respectively, with an av-erage degree of 5. Our approach finds complex, large-scale queries in seconds while maintaining comparable precision and recall to slower current approaches. 1.
Project-Team Pulsar Perception Understanding Learning Systems for Activity Recognition
"... c t i v it y e p o r t ..."
(Show Context)
Surveillance video retrieval: what we have already
"... Abstract—While many overview papers have been published for information retrieval in general and image retrieval in particular, there is a lack of paper in the literature focusing on retrieval for surveillance video. The aim of this paper is to provide an analysis on what we have ready done for surv ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract—While many overview papers have been published for information retrieval in general and image retrieval in particular, there is a lack of paper in the literature focusing on retrieval for surveillance video. The aim of this paper is to provide an analysis on what we have ready done for surveillance video retrieval and therefore to point out what are still challenges in this domain. By supposing that there are two main types of information in surveillance video named object and event, we divide the existing approaches in the literature into two sub categories: approaches at object level and approaches at both object and event levels. A quantitative comparison of three
Unsupervised Surveillance Video Retrieval based on Human Action and Appearance
"... Abstract—Forensic video analysis is the offline analysis of video aimed at understanding what happened in a scene in the past. Two of its key tasks are the recognition of specific actions, e.g., walking or fighting, and the search for specific persons, also referred to as re-identification. Although ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract—Forensic video analysis is the offline analysis of video aimed at understanding what happened in a scene in the past. Two of its key tasks are the recognition of specific actions, e.g., walking or fighting, and the search for specific persons, also referred to as re-identification. Although these tasks have traditionally been performed manually in forensic investigations, the current growing number of cameras and recorded video leads to the need for automated analysis. In this paper we propose an unsupervised retrieval system for surveillance videos based on human action and appearance. Given a query window, the system retrieves people performing the same action as the one in the query, the same person performing any action, or the same person performing the same action. We use an adaptive search algorithm that focuses the analysis on relevant frames based on the inter-frame difference of foreground masks. Then, for each analyzed frame, a pedestrian detector is used to extract windows containing each pedestrian in the scene. For each detection, we use optical flow features to represent its action and color features to represent its appearance. These extracted features are used to compute the probability that the detection matches the query according to the specified criterion. The algorithm is fully unsupervised, i.e., no training or constraints on the appearance, actions or number of actions that will appear in the test video are made. The proposed algorithm is tested on a surveillance video with different people performing different actions, providing satisfactory retrieval performance. I.
3 Appearance-Based Retrieval for Tracked Objects in
"... Video surveillance is a rapidly growing industry. Driven by low-hardware costs, heightened security fears and increased capabilities, video surveillance equipment is being deployed more widely and with greater storage than ever. This provides a huge amount of video data. Associating to these video d ..."
Abstract
- Add to MetaCart
Video surveillance is a rapidly growing industry. Driven by low-hardware costs, heightened security fears and increased capabilities, video surveillance equipment is being deployed more widely and with greater storage than ever. This provides a huge amount of video data. Associating to these video data, retrieval facilities become very useful for many purposes
