Abstract
Person re-identification (Re-id) is one of the important tools of video surveillance systems, which aims to recognize an individual across the multiple disjoint sensors of a camera network. Despite the recent advances on RGB camera-based person re-identification methods under normal lighting conditions, Re-id researchers fail to take advantages of modern RGB-D sensor-based additional information (e.g. depth and skeleton information). When traditional RGB-based cameras fail to capture the video under poor illumination conditions, RGB-D sensor-based additional information can be advantageous to tackle these constraints. This work takes depth images and skeleton joint points as additional information along with RGB appearance cues and proposes a person re-identification method. We combine 4-channel RGB-D image features with skeleton information using score-level fusion strategy in dissimilarity space to increase re-identification accuracy. Moreover, our propose method overcomes the illumination problem because we use illumination invariant depth image and skeleton information. We carried out rigorous experiments on two publicly available RGBD-ID re-identification datasets and proved the use of combined features of 4-channel RGB-D images and skeleton information boost up the rank 1 recognition accuracy.
Highlights
Person re-identification is the task of recognizing an individual across the distributed camera views
Our contributions are summarized as follows: 1. We propose a person re-identification method for fusing 4-channel RGB-D and skeleton information in dissimilarity space to gain high re-identification accuracy
PROPOSED APPROACH we describes our proposed skeleton and 4-channel RGB-D image-based person re-identification method
Summary
Person re-identification is the task of recognizing an individual across the distributed camera views. Imani et al [10] consider local shape descriptor and anthropometric measure to tackle the illumination problem Both methods use score-level fusion techniques to combine depth and skeleton features. Using 4-channel RGB-D images, we extract robust features using a deep CNN network Beside this features, we consider another anthropometric features, skeleton information of an individual, which are illumination and color invariant. Ren et al [13, 14] introduced a multi-modal uniform and variational deep learning method for person re-identification, where [13] define a network to combine features extracted from processed depth images and RGB images to a fusion layer. Imani et al [25] proposed a short term person re-identification method They extract local vector pattern both from RGB and depth modalities. We follow the batch-hard triplet generation strategy [35] for choosing right combination of triplet inputs
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
