Deep Hybrid Similarity Learning for Person Re-identification
This work addresses the problem of matching person images across non-overlapping cameras for surveillance and security applications, representing an incremental improvement in metric learning for person re-identification.
The paper tackled person re-identification by proposing a deep hybrid similarity learning method that combines element-wise absolute difference and multiplication of CNN features to improve matching accuracy. It achieved superior performance on three benchmark datasets, including QMUL GRID, VIPeR, and CUHK03, compared to state-of-the-art methods.
Person Re-IDentification (Re-ID) aims to match person images captured from two non-overlapping cameras. In this paper, a deep hybrid similarity learning (DHSL) method for person Re-ID based on a convolution neural network (CNN) is proposed. In our approach, a CNN learning feature pair for the input image pair is simultaneously extracted. Then, both the element-wise absolute difference and multiplication of the CNN learning feature pair are calculated. Finally, a hybrid similarity function is designed to measure the similarity between the feature pair, which is realized by learning a group of weight coefficients to project the element-wise absolute difference and multiplication into a similarity score. Consequently, the proposed DHSL method is able to reasonably assign parameters of feature learning and metric learning in a CNN so that the performance of person Re-ID is improved. Experiments on three challenging person Re-ID databases, QMUL GRID, VIPeR and CUHK03, illustrate that the proposed DHSL method is superior to multiple state-of-the-art person Re-ID methods.