Abstract
Person re-identification (Re-ID) is a challenging and arduous task due to non-overlapping views, complex background, and uncontrollable occlusion in video surveillance. An existing method for capturing pedestrian local region information is to divide person regions into horizontal stripes, which may lead to invalid features and erroneous learning. To solve this problem, this paper proposes a 3D skeleton and a two-stream approach to person Re-ID. The first stream of the method uses the 3D skeleton for background filtering and region segmentation. The second stream uses Siamese net to extract the global descriptor. The features of the two streams are fused to preserve the integrity of the person. An optimized region matching method for metric learning is designed. Extensive comparing experiments were conducted with state-of-the-art Re-ID methods on the Market-1501, CUHK03, and DukeMTMC-reID datasets. Experimental results show that the proposed method outperforms the existing methods in recognition accuracy.
- [1] . 2018. Sharp attention network via adaptive sampling for person re-identification. IEEE Trans. Circuits Syst. Video Technol. 29, 10 (2018), 3016–3027. Google Scholar
Cross Ref
- [2] . 2019. Progressive learning for person re-identification with one example. IEEE Trans. Image Process. 28, 6 (2019), 2872–2881. Google Scholar
Cross Ref
- [3] . 2021. Intra-inter camera similarity for unsupervised person re-identification. In 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR’21). IEEE, Nashville, TN, USA, 11926–11935. Google Scholar
Cross Ref
- [4] . 2017. Deep hybrid similarity learning for person re-identification. IEEE Trans. Circuits Syst. Video Technol. 28, 11 (2017), 3183–3193. Google Scholar
Digital Library
- [5] . 2020. Unsupervised person re-identification via cross-camera similarity exploration. IEEE Trans. Image Process. 29 (2020), 5481–5490. Google Scholar
Cross Ref
- [6] . 2020. Visible-infrared person re-identification via homogeneous augmented tri-modal learning. IEEE Trans. Inf. Forensic Secur. 16 (2020), 728–739. Google Scholar
Cross Ref
- [7] . 2020. Detecting motion blurred vehicle logo in IoV using Filter-DeblurGAN and VL-YOLO. IEEE Trans. Veh. Technol. 69, 4 (2020), 3604–3614. Google Scholar
Cross Ref
- [8] . 2019. Omni-scale feature learning for person re-identification. In 2019 IEEE/CVF International Conference on Computer Vision (CVPR’19). IEEE, Seoul, Korea (South), 3702–3712. Google Scholar
Cross Ref
- [9] . 2018. Fusion of multiple person re-id methods with model and data-aware abilities. IEEE T. Cybern. 50, 2 (2018), 561–571. Google Scholar
Cross Ref
- [10] . 2019. Learning what and where from attributes to improve person re-identification. In 2019 IEEE International Conference on Image Processing (ICIP’19). IEEE, Taipei, Taiwan, 165–169. Google Scholar
Cross Ref
- [11] . 2017. A discriminatively learned CNN embedding for person re-identification. ACM Trans. Multimed. Comput. Commun. 14, 1 (2017), 1–20. Google Scholar
Digital Library
- [12] . 2016. A Siamese long short-term memory architecture for human re-identification. In 2016 European Conference on Computer Vision (ECCV’16). Springer, Amsterdam, The Netherlands, Springer, Cham, 135–153. Google Scholar
Cross Ref
- [13] . 2016. An enhanced deep feature representation for person re-identification. In 2016 IEEE Winter Conference on Applications of Computer Vision (WACV’16). IEEE, Lake Placid, NY, USA, 1–8. Google Scholar
Cross Ref
- [14] . 2019. Pose-invariant embedding for deep person re-identification. IEEE Trans. Image Process 28, 9 (2019), 4500–4509. Google Scholar
Cross Ref
- [15] . 2018. GLAD: Global–local-alignment descriptor for scalable person re-identification. IEEE Trans. Multimedia 21, 4 (2018), 986–999. Google Scholar
Cross Ref
- [16] . 2016. Gated Siamese convolutional neural network architecture for human re-identification. In 2016 European Conference on Computer Vision (ECCV’16). Springer, Amsterdam, The Netherlands, Springer, Cham, 791–808. Google Scholar
Cross Ref
- [17] . 2019. OpenPose: Realtime multi-person 2D pose estimation using part affinity fields. IEEE Trans. Pattern Anal. Mach. Intell. 43, 1 (2019), 172–186. Google Scholar
Digital Library
- [18] . 2020. Driver yawning detection based on subtle facial action recognition. IEEE Trans. Multimedia 23, (2020), 572–583. Google Scholar
Cross Ref
- [19] . 2019. A bottom-up clustering approach to unsupervised person re-identification. In 2019 AAAI Conference on Artificial Intelligence (AAAI’19). AAAI, Hawaii, USA, 8738–8745. Google Scholar
Digital Library
- [20] . 2017. A new approach to track multiple vehicles with the combination of robust detection and two classifiers. IEEE Trans. Intell. Transp. Syst. 19, 1 (2017), 174–186. Google Scholar
Cross Ref
- [21] . 2020. STNReID: Deep convolutional networks with pairwise spatial transformer networks for partial person re-identification. IEEE Trans. Multimedia 22, 11 (2020), 2905–2913. Google Scholar
Cross Ref
- [22] . 2019. Person re-identification between visible and thermal camera images based on deep residual CNN using single input. IEEE Access 7 (2019), 57972–57984. Google Scholar
Cross Ref
- [23] . 2019. True-color and grayscale video person re-identification. IEEE Trans. Inf. Forensic Secur. 15 (2019), 115–129. Google Scholar
Cross Ref
- [24] . 2019. In defense of color names for small-scale person re-identification. In 2019 International Conference on Biometrics (ICB’19). IEEE, Crete, Greece, 1–6. Google Scholar
Cross Ref
- [25] . 2020. Discriminative fine-grained network for vehicle re-identification using two-stage re-ranking. Sci. China-Inf. Sci. 63, 11 (2020), 1–12. Google Scholar
Cross Ref
- [26] . 2015. Person re-identification by local maximal occurrence representation and metric learning. In 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’15). IEEE, Boston, MA, USA, 2197–2206. Google Scholar
Cross Ref
- [27] . 2019. Unsupervised person re-identification by soft multi label learning. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR’19). IEEE, Seoul, Korea (South), 2148–2157. Google Scholar
Cross Ref
- [28] . 2017. Joint detection and identification feature learning for person search. In 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’17). IEEE, Honolulu, HI, USA, 3415–3424. Google Scholar
Cross Ref
- [29] . 2014. DeepReID: Deep filter pairing neural network for person re-identification. In 2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’14). IEEE, Columbus, OH, USA, 152–159. Google Scholar
Digital Library
- [30] . 2016. PersonNet: Person re-identification with deep convolutional neural networks. (2016) arXiv:1601.07255. Retrieved from https://arxiv.org/abs/1601.07255Google Scholar
- [31] . 2019. Invariance matters: Exemplar memory for domain adaptive person re-identification. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR’19). IEEE, Seoul, Korea (South), 598–607. Google Scholar
Cross Ref
- [32] . 2020. SIF: Self-inspirited feature learning for person re-identification. IEEE Trans. Image Process. 29 (2020), 4942—4951. Google Scholar
Digital Library
- [33] . 2019. Improving person re-identification by attribute and identity learning. Pattern Recognit. 95 (2019), 151–161. Google Scholar
Digital Library
- [34] . 2019. Deep-person: Learning discriminative deep features for person re-identification. Pattern Recognit. 98 (2019), 107036. Google Scholar
Digital Library
- [35] . 2018. Beyond part models: Person retrieval with refined part pooling (and a strong convolutional baseline). In 2018 IEEE Conference on the European Conference on Computer Vision (CVPR’18). IEEE, Salt Lake City, UT, USA, 480–496. Google Scholar
Cross Ref
- [36] . 2016. Improving person re-identification via pose-aware multi-shot matching. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’16). IEEE, Las Vegas, NV, USA, 1354–1362. Google Scholar
Cross Ref
- [37] . 2018. Pedestrian alignment network for large-scale person re-identification. IEEE Trans. Circuits Syst. Video Technol. 29, 10 (2018), 3037–3045. Google Scholar
Cross Ref
- [38] . 2015. Mapping appearance descriptors on 3D body models for people re-identification. Int. J. Comput. Vis. 111, 3 (2015), 345–364. Google Scholar
Digital Library
- [39] . 2020. Feature mask network for person re-identification. Pattern Recognit. Letters 137 (2020), 91–98. Google Scholar
Cross Ref
- [40] . 2018. Harmonious attention network for person re-identification. In 2018 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’18). IEEE, Salt Lake City, UT, USA, 2285–2294. Google Scholar
Cross Ref
- [41] . 2019. Pose-guided feature alignment for occluded person re-identification. In 2019 IEEE/CVF International Conference on Computer Vision (ICCV’19). IEEE, Seoul, Korea (South), 542–551. Google Scholar
Cross Ref
- [42] . 2018. Dual attention matching network for context-aware feature sequence based person re-identification. In 2018 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’18). IEEE, Salt Lake City, UT, USA, 5363–5372. Google Scholar
Cross Ref
- [43] . 2015. Bilinear CNN models for fine-grained visual recognition. In 2015 IEEE International Conference on Computer Vision (ICCV’15). IEEE, Santiago, Chile, 1449–1457. Google Scholar
Digital Library
- [44] . 2005. ViPER: Augmenting automatic information extraction with visual perceptions. In Proceedings of the 14th ACM International Conference on Information and Knowledge Management (CIKM’05). ACM, Bremen, Germany, 381–388. Google Scholar
Digital Library
- [45] . 2018. An adaptive multi-projection metric learning for person re-identification across non-overlapping cameras. IEEE Trans. Circuits Syst. Video Technol. 29, 9 (2018), 2809–2821. Google Scholar
Digital Library
- [46] . 2018. Person re-identification using hybrid representation reinforced by metric learning. IEEE Access 6 (2018), 77334–77349. Google Scholar
Cross Ref
- [47] . 2016. Learning a discriminative null space for person re-identification. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’16). IEEE, Las Vegas, NV, USA, 1239–1248. Google Scholar
Cross Ref
- [48] . 2017. Re-ranking person re-identification with k-reciprocal encoding. In 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’17). IEEE, Honolulu, HI, USA, 1318–1327. Google Scholar
Cross Ref
- [49] . 2015. ImageNet large scale visual recognition challenge. Int. J. Comput. Vis. 115, 3 (2015), 211–252. Google Scholar
Digital Library
- [50] . 2017. SVDNet for pedestrian retrieval. In 2017 IEEE International Conference on Computer Vision (ICCV’17). IEEE, Venice, Italy, 3800–3808. Google Scholar
Cross Ref
- [51] . 2015. Scalable person re-identification: A benchmark. In 2015 IEEE International Conference on Computer Vision (ICCV’15). IEEE, Santiago, Chile, 1116–1124. Google Scholar
Cross Ref
- [52] . 2016. Performance measures and a data set for multi-target, multi-camera tracking. In 2016 European Conference on Computer Vision (ECCV’16). Springer, Amsterdam, The Netherlands, 17–35. Google Scholar
Cross Ref
- [53] . 2017. In defense of the triplet loss for person re-identification. (2017), arXiv:1703.07737. Retrieved from https://arxiv.org/abs/1703.07737Google Scholar
- [54] . 2018. Squeeze-and-excitation networks. In 2018 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’18). IEEE, Salt Lake City, UT, USA, 7132–7141. Google Scholar
Digital Library
- [55] . 2017. Person re-identification with metric learning using privileged information. IEEE Trans. Image Process. 27, 2 (2017), 791–805. Google Scholar
Cross Ref
- [56] . 2017. Enhancing person re-identification in a self-trained subspace. ACM Trans. Multimed. Comput. Commun. 13, 3 (2017), 1—23. Google Scholar
Digital Library
- [57] . 2012. Multimodal graph-based reranking for web image search. IEEE Trans. Image Process. 21, 11 (2012), 4649–4661. Google Scholar
Digital Library
- [58] . 2020. Learning to match on graph for fashion compatibility modeling. In 2020 AAAI Conference on Artificial Intelligence (AAAI’20). AAAI, New York, NY, USA, 287–294. Google Scholar
Cross Ref
- [59] . 2019. Improved person re-identification based on saliency and semantic parsing with deep neural network models. Image Vis. Comput. 92 (2019), 103809. Google Scholar
Digital Library
- [60] . 2018. FD-GAN: Pose-guided feature distilling GAN for robust person re-identification. NIPS, Montréal, Canada. arXiv:1810.02936. Retrieved from https://arxiv.org/abs/1810.02936Google Scholar
- [61] . 2018. Part-aligned bilinear representations for person re-identification. In Proceedings of European Conference on Computer Vision (ECCV’18). Springer, Munich, Germany, 402–419. Google Scholar
Cross Ref
- [62] . 2018. Multi-level factorisation net for person re-identification. In 2018 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’18). IEEE, Salt Lake City, UT, USA, 2109–2118. Google Scholar
Cross Ref
Index Terms
3D Skeleton and Two Streams Approach to Person Re-identification Using Optimized Region Matching
Recommendations
Local region partition for person re-identification
AbstractDue to the different posture and view angle, the image will appear some objects that do not exist in another image of the same person captured by another camera. The region covered by new items adversely improved the difficulty of person re-...
Person re-identification using salient region matching game
The human saliency feature has been increasingly used for person re-identification across non-overlapping cameras but is deficient in retaining the minor features of the salient region, thus resulting in matching accuracy decline. To address this ...
Person Re-identification Using Region Covariance in a Multi-feature Approach
Image Analysis and RecognitionAbstractPerson re-identification is an important requirement for modern video surveillance systems and relevant for human tracking, especially over camera networks. Many different approaches have been proposed but a robust identification under real-life ...






Comments