Re-Identification of person aims at retrieval of person across multiple non overlapping camera. There was a huge gain in the computer vision community with the advancement of deep learning features and also the number of surveillance in videos increased. The challenges faced by person re-identification is low resolution images, pose variation etc., and convolutional neural networks are supported by a number of state-of-the-art algorithms for person re-identification. In this paper, Siamese network is used to predict the similarity or dissimilarity of a person across two cameras. It's a neural architecture that takes as input a pair of images or videos and the output as the prediction of similar and dissimilar persons along with their prediction scores. The experimentation is done by using datasets iLIDS-VID, PRID 2011 and obtained a recognition accuracy of 79.52% and 85.82% respectively.

No abstract

Age and gender are the two key attributes for healthy social interactions, access control, intelligence marketing etc. Likewise, carried object recognition helps in identifying owner of the baggage being abandoned or the person littering in the public places. The above‐mentioned surveillance task displays discriminative characteristics in gait. Primates can accomplish scene context understanding and reacting to different circumstances with varying reflexes with ease. Human beings achieve this by recollecting prior experiences and adapting to new situations quickly. Modelling the human behaviour, this research work has combined customized and learnable filters so that knowledge database can always be kept up to date, as well as, provides flexibility in learning new contexts. Thus, a specialized parallel deep convolutional neural network architecture with customized filters that extracts intrinsic characteristics and data driven learnable filters are fused to enhance the performance of single convolutional neural network is proposed. From the experimentation it is observed that, the learning is augmented when customized filters and learnable filters are fused together. Results show that the proposed system achieves better performance for CASIA B datAQ2abase and OU‐ISIR gait database‐large population dataset with age and real‐life carried object.

No abstract

No abstract