A New System for Event Detection from Video Surveillance Sequences

Human action recognition based on multi-layer Fisher vector encoding method

“…Human action recognition is becoming one of the most important topics in video processing due to the numerous applications such as video-surveillance (Wali et al, 2010), human computer interaction (Bouchrika et al, 2013), video indexing and retrieving (Mejdoub et al, 2009), (Ben Aoun et al, 2011b), (Ben Aoun et al, 2011a) etc. Having no previous knowledge about the location of the person in each video frame, the human action in a video stream can be recovered from a great number of local descriptors extracted from the video frames (Sekma et al, 2013), (Dammak et al, 2012), , (Sekma et al, 2014).…”

Fast beta wavelet network-based feature extraction for image copy detection

“…Second, it is used for learning. This network, which is a neural network in which the hidden layer contains wavelet functions, reduces computational costs, considerably, and it was an effective tool in various fields such as image compression [26], image retrieval [27][28][29], face recognition [30], image classification [31,32], speech recognition [33] and in particular arabic word recognition [34], object tracking and recognition [35][36][37].…”

Person re-ID while Crossing Different Cameras: Combination of Salient-Gaussian Weighted BossaNova and Fisher Vector Encodings

“…The person re-identification (re-ID) [1,2,3,4,5,6,7,8,9] goal is to retrieve gallery images containing the same person as the probe (query) in a video surveillance [10,11] crosscamera mode. In general, images containing human subjects who are pre-captured by a detector or a tracker, form the input of a person re-ID system.…”

“…Indeed, our work is specially motivated by studying the effectiveness of the histogram encoding methods [23] in the person re-ID task, since they can simplify the matching process between persons, with respect to the brute force matching methods. Bag Of Statistical Sampling Analysis BossaNova (BN) [24,25] is a pooling method that demonstrated its effectiveness in the image classification field [11,26,27,28,29,30,31,32,33,34]. It consists in discretizing the patches to clusters' assignments into several Bins.…”