Paper Search Console

Home Search Page About Contact

Journal Title

Title of Journal: Int J Comput Vis

Search In Journal Title:

Abbravation: International Journal of Computer Vision

Search In Journal Abbravation:

Publisher

Springer US

Search In Publisher:

DOI

10.1016/0550-3213(83)90652-1

Search In DOI:

ISSN

1573-1405

Search In ISSN:
Search In Title Of Papers:

ExemplarGuided Similarity Learning on Polynomial

Authors: Dapeng Chen Zejian Yuan Jingdong Wang Badong Chen Gang Hua Nanning Zheng
Publish Date: 2017/02/13
Volume: 123, Issue: 3, Pages: 392-414
PDF Link

Abstract

Person reidentification is a crucial problem for video surveillance aiming to discover the correct matches for a probe person image from a set of gallery person images To directly describe the image pair we present a novel organization of polynomial kernel feature map in a high dimensional feature space to break down the variability of positive person pairs An exemplarguided similarity function is built on the map which consists of multiple subfunctions Each subfunction is associated with an “exemplar” image being responsible for a particular type of image pair thus excels at separating the persons with similar appearance We formulate a unified learning problem including a relaxed loss term as well as two kinds of regularization strategies particularly designed for the feature map The corresponding optimization algorithm jointly optimizes the coefficients of all the subfunctions and selects the proper exemplars for a better discrimination The proposed method is extensively evaluated on six public datasets where we thoroughly analyze the contribution of each component and verify the generalizability of our approach by crossdataset experiments Results show that the new method can achieve consistent improvements over stateoftheart methodsThis work was supported by the National Key Research and Development Program of China No 2016YFB1001001 the National Basic Research Program of China No 2015CB351703 No 2012CB316400 the National Natural Science Foundation of China No 61573280 No 91648121 No 61603022 No 61573273

Keywords:

References

.
Search In Abstract Of Papers:
Other Papers In This Journal:
  1. Growing Regression Tree Forests by Classification for Continuous Object Pose Estimation
  2. Objects, Actions, Places
  3. Hierarchical Shape Segmentation and Registration via Topological Features of Laplace-Beltrami Eigenfunctions
  4. Geometric Image Parsing in Man-Made Environments
  5. On Taxonomies for Multi-class Image Categorization
  6. Full and Partial Symmetries of Non-rigid Shapes
  7. A Two-Layer Framework for Piecewise Linear Manifold-Based Head Pose Estimation
  8. Learning Discriminative Space–Time Action Parts from Weakly Labelled Videos
  9. Recognizing Fine-Grained and Composite Activities Using Hand-Centric Features and Script Data
  10. 3D Human Pose Tracking Priors using Geodesic Mixture Models
  11. Ethnicity- and Gender-based Subject Retrieval Using 3-D Face-Recognition Techniques
  12. A Variational Approach to Video Registration with Subspace Constraints
  13. The Fisher-Rao Metric for Projective Transformations of the Line
  14. Planar Motion Estimation and Linear Ground Plane Rectification using an Uncalibrated Generic Camera
  15. Guest Editorial: Human Activity Understanding from 2D and 3D Data
  16. Fast and Stable Polynomial Equation Solving and Its Application to Computer Vision
  17. Robust Pose Recognition of the Obscured Human Body
  18. Teichmüller Shape Descriptor and Its Application to Alzheimer’s Disease Study
  19. Parsing Images into Regions, Curves, and Curve Groups
  20. Using Biologically Inspired Features for Face Processing
  21. Two-View Motion Segmentation with Model Selection and Outlier Removal by RANSAC-Enhanced Dirichlet Process Mixture Models
  22. Information-Theoretic Active Polygons for Unsupervised Texture Segmentation
  23. Virtual Volumetric Graphics on Commodity Displays Using 3D Viewer Tracking
  24. 3-D Symmetry Detection and Analysis Using the Pseudo-polar Fourier Transform
Search Result: