Two Stream Deep CNN-RNN Attentive Pooling Architecture for Video-Based Person Re-identification

Person re-identification (re-ID), is the task of associating the relationship among the images of a person captured from different cameras with non-overlapping field of view. Fundamental and yet an open issue in re-ID is extraction of powerful features in low resolution surveillance videos. In order to solve this, a novel Two Stream Convolutional Recurrent model with Attentive pooling mechanism is presented for person re-ID in videos. Each stream of the model is a Siamese network which is aimed at extracting and matching most differentiated feature maps. Attentive pooling is used to select most informative video frames. The output of two streams is fused to formulate one combined feature map, which helps to deal with major challenges of re-ID e.g. pose and illumination variation, clutter background and occlusion. The proposed technique is evaluated on three challenging datasets: MARS, PRID-2011 and iLIDS-VID. Experimental evaluation shows that the proposed technique performs better than existing state-of-the-art supervised video based person re-ID models. The implementation is available at https://github.com/re-identification/Person_RE-ID.git.

[1]  Muhammad Moazam Fraz,et al.  Detailed Sentence Generation Architecture for Image Semantics Description , 2018, ISVC.

[2]  Takeo Kanade,et al.  An Iterative Image Registration Technique with an Application to Stereo Vision , 1981, IJCAI.

[3]  Qi Tian,et al.  MARS: A Video Benchmark for Large-Scale Person Re-Identification , 2016, ECCV.

[4]  Bingpeng Ma,et al.  A Spatio-Temporal Appearance Representation for Video-Based Pedestrian Re-Identification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[5]  Horst Bischof,et al.  Person Re-identification by Descriptive and Discriminative Classification , 2011, SCIA.

[6]  Muhammad Moazam Fraz,et al.  DUPL-VR: Deep Unsupervised Progressive Learning for Vehicle Re-Identification , 2018, ISVC.

[7]  Yang Li,et al.  Sparse re-id: Block sparsity for person re-identification , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[8]  Muhammad Moazam Fraz,et al.  Optimization of Person Re-Identification through Visual Descriptors , 2018, VISIGRAPP.

[9]  Edward J. Delp,et al.  A Two Stream Siamese Convolutional Neural Network for Person Re-identification , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[10]  Shaogang Gong,et al.  Person Re-identification by Video Ranking , 2014, ECCV.

[11]  Shengcai Liao,et al.  Deep Metric Learning for Person Re-identification , 2014, 2014 22nd International Conference on Pattern Recognition.

[12]  Deqiang Ouyang,et al.  Video-based person re-identification via spatio-temporal attentional and two-stream fusion convolutional networks , 2019, Pattern Recognit. Lett..

[13]  Muhammad Moazam Fraz,et al.  Person Re-Identification Using Hybrid Representation Reinforced by Metric Learning , 2018, IEEE Access.

[14]  Yu Cheng,et al.  Jointly Attentive Spatial-Temporal Pooling Networks for Video-Based Person Re-identification , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[15]  Muhammad Moazam Fraz,et al.  Weighted hybrid features for person re-identification , 2017, 2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA).

[16]  Jesús Martínez del Rincón,et al.  Recurrent Convolutional Network for Video-Based Person Re-identification , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[17]  Ziyan Wu,et al.  A Systematic Evaluation and Benchmark for Person Re-Identification: Features, Metrics, and Datasets , 2016, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[18]  Bernd Girod,et al.  Recurrent Neural Networks for Person Re-identification Revisited , 2019, 2019 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR).