Search

We describe the TNO system and the evaluation results for TRECVID 2013 Multimedia Event Detection (MED) and instance search (INS) tasks. The MED system consists of a bag-of-word (BOW) approach with spatial tiling that uses low-level static and dynamic visual features, an audio feature and high-level concepts. Automatic speech recognition (ASR) and optical character recognition (OCR) are not used in the system. In the MED case with 100 example training videos, support-vector machines (SVM) are trained and fused to detect an event in the test set. In the case with 0 example videos, positive and negative concepts are extracted as keywords from the textual event description and events are detected with the high-level concepts. The MED results show that the SIFT keypoint descriptor is the one which contributes best to the results, fusion of multiple low-level features helps to improve the performance, and the textual event-description chain currently performs poorly. The TNO INS system presents a baseline open-source approach using standard SIFT keypoint detection and exhaustive matching. In order to speed up search times for queries a basic map-reduce scheme is presented to be used on a multi-node cluster. Our INS results show above-median results with acceptable search times., This research for the MED submission was performed in the GOOSE project, which is jointly funded by the enabling technology program Adaptive Multi Sensor Networks (AMSN) and the MIST research program of the Dutch Ministry of Defense. The INS submission was partly supported by the MIME project of the creative industries knowledge and innovation network CLICKNL., peer-reviewed

More and more sensors are getting Internet connected. Examples are cameras on cell phones, CCTV cameras for traffic control as well as dedicated security and defense sensor systems. Due to the steadily increasing data volume, human exploitation of all this sensor data is impossible for effective mission execution. Smart access to all sensor data acts as enabler for questions such as “Is there a person behind this building” or “Alert me when a vehicle approaches”. The GOOSE concept has the ambition to provide the capability to search semantically for any relevant information within “all” (including imaging) sensor streams in the entire Internet of sensors. This is similar to the capability provided by presently available Internet search engines which enable the retrieval of information on “all” web pages on the Internet. In line with current Internet search engines any indexing services shall be utilized cross-domain. The two main challenge for GOOSE is the Semantic Gap and Scalability. The GOOSE architecture consists of five elements: (1) an online extraction of primitives on each sensor stream; (2) an indexing and search mechanism for these primitives; (3) a ontology based semantic matching module; (4) a top-down hypothesis verification mechanism and (5) a controlling man-machine interface. This paper reports on the initial GOOSE demonstrator, which consists of the MES multimedia analysis platform and the CORTEX action recognition module. It also provides an outlook into future GOOSE development.