본문 바로가기
KERD
논문KCI 등재

Analyzing Pedestrian Parts using Deep Features for Person Re-Identification

이미영|백성욱|라힘 나시르|아마드 자밀|울라 아민
한국차세대컴퓨팅학회 (KINGComputing)|2017.04|한국차세대컴퓨팅학회 논문지 (The Journal of Korean Institute of Next Generation Computing)|제13권 제2호|pp. 82-92

2

조회

0

다운로드

4

인용

0

참고문헌

키워드

이미지검색Image retrievalDeep featuresPerson re-identificationContent-based search딥 기능사람 재식별콘텐츠 기반 검색
주제분류컴퓨터학

초록

In computer vision, person re-identification is a challenging task in the context of making intelligent surveillance system. State-of-the-art techniques in person re-identification systems still rely on traditional features extraction such as HOG, SIFT, SURF, and LBP etc. In this study we present convolution neural network (CNN) features to represent persons in surveillance streams. Firstly, we trained a CNN model using person re-id dataset. Secondly, we divide pedestrian images into upper and lower body parts, keeping in view the format of clothing used to cover upper and lower parts. Further, we have extracted CNN features for both parts using the trained model. Similarity score is computed as Euclidean distance between query image and other images in the dataset. We succeeded to achieve higher precision scores on various recall levels, as well as higher Cumulative Match Characteristic (CMC) performance scores compared to other state-of-the-art techniques on ETHZ, CUHK, and Market datasets.

Abstract

In computer vision, person re-identification is a challenging task in the context of making intelligent surveillance system. State-of-the-art techniques in person re-identification systems still rely on traditional features extraction such as HOG, SIFT, SURF, and LBP etc. In this study we present convolution neural network (CNN) features to represent persons in surveillance streams. Firstly, we trained a CNN model using person re-id dataset. Secondly, we divide pedestrian images into upper and lower body parts, keeping in view the format of clothing used to cover upper and lower parts. Further, we have extracted CNN features for both parts using the trained model. Similarity score is computed as Euclidean distance between query image and other images in the dataset. We succeeded to achieve higher precision scores on various recall levels, as well as higher Cumulative Match Characteristic (CMC) performance scores compared to other state-of-the-art techniques on ETHZ, CUHK, and Market datasets.

참고문헌 (48)

  1. Ahmad, J., "Efficient object-based surveillance image search using spatial pooling of convolutional features", Journal of Visual Communication and Image Representation, 45, pp.62~76 (2017)
  2. Swain, M.J., "Indexing via color histograms, in Active Perception and Robot Vision", pp.261~273 (1992), Springer
  3. Bazzani, L., "Multiple-shot person reidentification by hpe signature", in Pattern Recognition (ICPR), 2010 20th International Conference on (2010), IEEE
  4. Corvee, E., "Person re-identification using spatial covariance regions of human body parts", in Advanced Video and Signal Based Surveillance (AVSS), 2010 Seventh IEEE International Conference on (2010), IEEE
  5. Bak, S., "Multiple-shot human reidentification by mean riemannian covariance grid", in Advanced Video and Signal-Based Surveillance (AVSS), 2011 8th IEEE International Conference on (2011), IEEE
  6. Bauml, M., "Evaluation of local features for person re-identification in image sequences", in Advanced Video and Signal-Based Surveillance (AVSS), 2011 8th IEEE International Conference on (2011), IEEE
  7. Farenzena, M., "Person re-identification by symmetry-driven accumulation of local features", in Computer Vision and Pattern Recognition (CVPR), 2010 IEEE Conference on (2010), IEEE
  8. Gheissari, N., "Person reidentification using spatiotemporal appearance", in Computer Vision and Pattern Recognition, 2006 IEEE Computer Society Conference on (2006), IEEE
  9. Gray, D., "Viewpoint invariant pedestrian recognition with an ensemble of localized features", in European conference on computer vision. 2008 (2008), Springer
  10. Wang, X., "Shape and appearance context modeling", in Computer Vision, 2007. ICCV 2007. IEEE 11th International Conference on (2007), IEEE
  11. Zhang, Y., "LBP based region covariance descriptor for person re-identification", in Image and Graphics (ICIG), 2011 Sixth International Conference on (2011), IEEE
  12. Bazzani, L., "Symmetry-driven accumulation of local features for human characterization and re-identification", Computer Vision andImage Understanding, 117(2), pp.130~144 (2013)
  13. Ma, B., "Covariance descriptor based on bio-inspired features for person re-identification and face verification", Image and Vision Computing, 32(6), pp.379~390 (2014)
  14. Zhao, R., "Learning mid-level filters for person re-identification", in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2014)
  15. Zhao, R., "Person re-identification by salience matching", in Proceedings of the IEEE International Conference on Computer Vision (2013)
  16. Dikmen, M., "Pedestrian recognition with a learned metric", in Asian conference on Computer vision. 2010 (2010), Springer
  17. Hu, Y., "Exploring structural information and fusing multiple features for person re-identification", in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops (2013)
  18. Hamdoun, O., "Person re-identification in multi-camera system by signature based on interest point descriptors collected on short video sequences", in Distributed Smart Cameras, 2008. ICDSC 2008. Second ACM, 2008, ~[DOI]
  19. Cheng, D.S., "Custom Pictorial Structures for Re-identification", Bmvc, ~ (2011)
  20. Sivic, J., "Video google: A text retrieval approach to object matching in videos", ICCV 2003 9th IEEE International Conference on Computer Vision (2003)
  21. Madden, C., "Tracking people across disjoint camera views by an illumination-tolerant appearance representation", Machine Vision and Applications, 18(3), pp.233~247 (2007)
  22. Prosser, B., "Multicamera matching under illumination change over time", in Workshop on Multi-camera and Multi-modal Sensor Fusion Algorithms and Applications-M2SFA2 2008 (2008)
  23. Prosser, B., "Person Re-Identification by Support Vector Ranking", BMVC, ~ (2010)
  24. Wang, H., "Effective appearance model and similarity measure for particle filtering and visual tracking", in European Conference on Computer Vision. 2006 (2006), Springer
  25. Ma, B., "Local descriptors encoded by fisher vectors for person reidentification", in European Conference on Computer Vision. 2012 (2012), Springer
  26. Zhao, R., "Unsupervised salience learning for person re-identification", in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2013)
  27. Liu, Y., "Person re-identification based on visual saliency", in Intelligent Systems Design and Applications (ISDA), 2012 12th International Conference on (2012), IEEE
  28. Wang, X., "Regionlets for generic object detection", in Proceedings of the IEEE Inter national Conference on Computer Vision (2013)
  29. Davis, J.V., "Information-theoretic metric learning", in Proceedings of the 24th international conference on Machine learning (2007), ACM
  30. Guillaumin, M., "Is that you? Metric learning approaches for face identification", in Computer Vision, 2009IEEE 12th international conference on (2009), IEEE
  31. Zheng, W.-S., "Person re-identification by probabilistic relative distance comparison", in Computer vision and pattern recognition (CVPR), 2011 IEEE conference on (2011), IEEE
  32. Koestinger, M., "Large scale metric learning from equivalence constraints", in Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on (2012), IEEE
  33. Hirzer, M., "Relaxed pairwise learned metric for person re-identification", in European Conference on Computer Vision (2012), Springer
  34. Li, Z., "Learning locally-adaptive decision functions for person verification", in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2013)
  35. Ahmad, J., "Saliency-weighted graphs for efficient visual content description and their applications in real-time image retrieval systems", Journal of Real-Time Image Processing, pp.1~17 (2016)
  36. Guo, Z., "A completed modeling of local binary pattern operator for texture classification", IEEE Transactions on Image Processing, 19(6), pp.1657~1663 (2010)
  37. Dalal, N., "Histograms of oriented gradients for human detection", in Computer Vision and Pattern Recognition, 2005. CVPR 2005. IEEE Computer Society Conference on (2005), IEEE
  38. Ng, P.C., "SIFT: Predicting amino acid changes that affect protein function", Nucleic acids research, 31(13), pp.3812~3814 (2003)
  39. Krizhevsky, A., "Imagenet classification with deep convolutional neural networks", in Advances in neural information processing systems (2012), Curran Associates, Inc
  40. Russakovsky, O., "Imagenet large scale visual recognition challenge", International Journal of Computer Vision, 115(3), pp.211~252 (2015)
  41. Schwartz, W.R., "Learning discriminative appearance-based models using partial least squares", in Computer Graphics and Image Processing (SIBGRAPI), 2009XXII Brazilian Symposium on (2009), IEEE
  42. Ahmad, J., "Multi-scale local structure patterns histogram for describing visual contents in social image retrieval systems", Multimedia Tools and Applications, 75(20), pp.12669~12692 (2016)
  43. Zheng, L., "Scalable person reidentification:A benchmark", in Proceedings of the IEEE International Conference on Computer Vision (2015)
  44. Li, W., "Deepreid: Deep filter pairing neural network for person re-identification", in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2014)
  45. Paisitkriangkrai, S., "Learning to rank in person reidentification with metric ensembles", in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2015)
  46. 박현철, "CNN에 기반한 실제 환경에서의 색상 및 문자 인식", 한국차세대컴퓨팅학회 논문지, 12(6), pp.104~115 (2016), 한국차세대컴퓨팅학회
  47. 박은성, "스마트 케어 감시 시스템에서 두 대의 CCTV를 이용한 GPS 위치정보의 보정 방법", 한국차세대컴퓨팅학회 논문지, 11(1), pp.54~63 (2015), 한국차세대컴퓨팅학회
  48. 성동수, "CCTV 카메라의 상태 정보에 근거하여 다수의 위치 정보를 처리하는 응급 상황 신고 서비스", 한국차세대컴퓨팅학회 논문지, 11(6), pp.51~59 (2015), 한국차세대컴퓨팅학회

관련 논문AI추천

A Study on Person Re-identification in Multi-view Surveillance Videos파트 유 민 울라, 김민제, 백성욱, 이미영, 이수민, 이자즈 울 하크한국차세대컴퓨팅학회 · 한국차세대컴퓨팅학회 논문지2021 · 제17권 제1호Gait Recognition Based on GF-CNN and Metric LearningJunqin Wen한국정보처리학회 · JIPS(Journal of Information Processing Systems)2020 · 제16권 제5호A Tracking-by-Detection System for Pedestrian Tracking Using Deep Learning Technique and Color Information김상훈, MAI THANH NHAT TRUONG한국정보처리학회 · JIPS(Journal of Information Processing Systems)2019 · 제15권 제4호The Improved Joint Bayesian Method for Person Re-identification Across Different CameraJiangtao Cao, Ligang Hou, Yingqiang Guo한국정보처리학회 · JIPS(Journal of Information Processing Systems)2019 · 제15권 제4호Human Identification Based on Gait Representation and AnalysisGang Yu, Yanghua Gao, Yuxiang Shan한국정보처리학회 · JIPS(Journal of Information Processing Systems)2024 · 제20권 제6호Person Identification based on Clothing Feature김구진, 박선미, 조위덕, 최유주(사)한국컴퓨터그래픽스학회 · 한국컴퓨터그래픽스학회논문지2010 · 제16권 제1호

이 간행물의 다른 논문 (9건)

Design and Implementation of an Automated Privacy Protection System over TPM and File Virtualization안성규, 박기웅, 김문성, 정혜림pp. 7-17오류가 발생한 멀티바이트 인코딩 데이터의 인코딩 기법 판별 알고리즘 개선김선범, 배준우, 박희진pp. 18-25Ensembles of Decision Trees Based Diagnosis of Alzheimer’s Disease and Dementia권구락, 데베쉬 자pp. 26-34스마트 환경에서의 사용자 상황인지 기반 지식필터링을 이용한 콘텐츠 추천 시스템염근혁, 김웅수, 이동우pp. 35-48멀티코어 환경에서 샘플링 기반 재시도 정책을이용한 하이브리드 트랜잭셔널 메모리장연우, 장재우, 윤민, 강문환pp. 49-61
원문 보기