Abstract

In this paper we propose a novel recurrent neural net-work architecture for video-based person re-identification.Given the video sequence of a person, features are extractedfrom each frame using a convolutional neural network thatincorporates a recurrent final layer, which allows informa-tion to flow between time-steps. The features from all time-steps are then combined using temporal pooling to give anoverall appearance feature for the complete sequence. Theconvolutional network, recurrent layer, and temporal pool-ing layer, are jointly trained to act as a feature extractor fovideo-based re-identification using a Siamese network ar-chitecture. Our approach makes use of colour and opticalflow information in order to capture appearance and motion information which is useful for video re-identification. Ex-periments are conduced on the iLIDS-VID and PRID-2011 datasets to show that this approach outperforms existingmethods of video-based re-identification.