http://www.journals.elsevier.com/pattern-recognition-letters/call-for-papers/special-issue-on-efficient-shape-representation-matching/
Pattern Recognition Letters
Special issue on Efficient Shape Representation, Matching, Ranking, and its Applications
As the era of big data is coming, efficiently and effectively representing, matching and ranking shape in large scale has become a crucial issue. Even though shape representation has been extensively researched, it has seldom been fully studied in large scale and still remains a hot topic in computer vision due to the fact that shape is a primary feature used by the human perception system to detect and identify objects. As a consequence it will continuously get much attention in many real-world applications of pattern recognition. This special issue will feature original research papers related to the theory, methods and algorithms for large scale shape representation, matching and ranking, together with applications to real-world problems.
Main Topics of interest: Theory and methods for shape matching: 1) Shape matching using any forms of representations including newly designed handcraft features, representations learned from shape data, graph-based models, etc in two or three dimensions. 2) Efficient shape matching\registration algorithms for establishing object correspondence. 3) Partial shape matching. 4) Shape recognition in supervised/unsupervised manner: shape classification/shape clustering. 5) Sketch matching/ classifcation.
Shape ranking/reranking theory and algorithm: Shape similarity measure, shape search and retrieval, shape learning to rank, contextual shape similarity: 1) new algorithms to explore the contexts from shape database for enhancing/reranking the results of shape retrieval/classification. 2) Fusing multiple shape representations for accurate object search.
Shape representation and related algorithms: Curvature, keypoint detection of contour/surface, skeleton/medial axis extraction and pruning, shape decomposition/segmentation, shape orientation, shape recovery, symmetry detection, shape representation using deep learning, etc
Shape-based object recognition: Object detection/recognition, contour detection/grouping, object segmentation using shape cues.
Applications: Application of shape matching to solve any real-world image understanding problems including object recognition in depth images, gesture/pose recognition/identification, biomedical imaging, document image analysis, shape modeling/animation, robot vision, geosciences and remote sensing, industrial imaging, etc. Submission guidelines: Authors should prepare their manuscript according to the Guide for Authors available on the online submission page of Pattern Recognition Letters. The authors must select as “SI: SrmrApp” when they reach the “Article Type” step in the submission step. Key deadlines/dates:
Submission deadline: Oct. 17, 2015
First review notification: Dec. 30, 2015
Revised submission due: Jan. 20, 2016
Notification of second-round review: Feb. 20, 2016
Second revised submission due (if necessary): March 10, 2016
Final notice of acceptance/rejection: March 25, 2016
For more information, please contact the Managing Guest Editor. Guest-Editors: Xiang Bai, Managing Guest Editor
School of Electronics Information and Communications, Huazhong University of Science and Technology, Wuhan, China 430074
Email: xbai@hust.edu.cn Michael Donoser, Guest Editor
Amazon Computer Vision Team Berlin, Amazon Development Center Germany, Kurfürstendamm 195, 10707 Berlin, Germany
Email: donoserm@amazon.com Hairong Liu, Guest Editor
Spansion, San Jose, USA, 94085
Email: lhrbss@gmail.com Longin Jan Latecki, Guest Editor
Department of Computer and Information Sciences, Temple University, Philadelphia, PA 19122, USA
Email: latecki@temple.edu
| 
发表于 2015-8-10 23:47:59
发表于 2015-9-27 11:08:04