3,800 Hits in 6.7 sec

Fast 3D Face Alignment and Improved Recognition Through Pyramidal Normal map Metric

Andrea F. Abate, Michele Nappi, Stefano Ricciardi, Gabriele Sabatino
2007 Proceedings of IEEE international conference on image processing  
This paper presents an improved normal map based face recognition approach, which relies on a novel method to automatically align a captured 3D face mesh to a reference template, allowing a more precise  ...  The alignment approach avoids the need for a rough or manual face pre-alignment and maximizes recognition precision, requiring a fraction of the time needed by the Iterative Closest Point (ICP) method  ...  CONCLUSIONS AND ONGOING RESEARCH We presented a face recognition method based on normal map metric and improved through a face registration procedure exploiting a novel iterative pyramidal-normal-map based  ... 
doi:10.1109/icip.2007.4378915 dblp:conf/icip/AbateNRS07 fatcat:ah4fvnox2fg4dpcjvo47ssa7ra

ABCNet: Real-time Scene Text Spotting with Adaptive Bezier-Curve Network [article]

Yuliang Liu, Hao Chen, Chunhua Shen, Tong He, Lianwen Jin, Liangwei Wang
2020 arXiv   pre-print
These methods either are costly for character annotation or need to maintain a complex pipeline, which is often not suitable for real-time applications.  ...  Our contributions are three-fold: 1) For the first time, we adaptively fit arbitrarily-shaped text by a parameterized Bezier curve. 2) We design a novel BezierAlign layer for extracting accurate convolution  ...  Acknowledgements The authors would like to thank Huawei Technologies for the donation of GPU cloud computing resources.  ... 
arXiv:2002.10200v2 fatcat:ef7lsgcflfaqxcqsv7rjisfazu

Spin Images and Neural Networks for Efficient Content-Based Retrieval in 3D Object Databases [chapter]

Pedro A. de Alarcón, Alberto D. Pascual-Montano, José M. Carazo
2002 Lecture Notes in Computer Science  
We describe a system for querying 3D model databases using the spin image representation as a shape signature for objects depicted as triangular meshes.  ...  The indexing schema improves significantly the efficiency in matching query spin images against those stored in the database.  ...  To do so, different features are used to represent the content of a 3D object. Among them, shape is the most relevant attribute for the task of object recognition.  ... 
doi:10.1007/3-540-45479-9_24 fatcat:igbk77dh2fguxge4to3mcvetaq

Circularity Measuring in Linear Time

Thanh Phuong Nguyen, Isabelle Debled-Rennesson
2010 2010 20th International Conference on Pattern Recognition  
We then introduce a linear algorithm that uses this measure for circularity measuring. This method can also be regarded as a method for circular object recognition.  ...  We propose a new circularity measure inspired from Arkin [1], Latecki [9] tools of shape matching that is constructed in a tangent space.  ...  Similarity of polygonal shapes Arkin et al. proposed a method based on L 2 distance [1] as a metric for comparing polygonal shapes. Later, Latecki et al.  ... 
doi:10.1109/icpr.2010.1157 dblp:conf/icpr/NguyenD10a fatcat:3dn27yphf5cfvhd473an5vkz2q

A Content-based Image Retrieval System Based On Convex Hull Geometry

2014 Acta Polytechnica Hungarica  
Convex hull shape polygon is extracted from the database images and the coordinate values are stored in the feature library.  ...  ) is used.  ...  We have proposed and implemented the CHAR method for efficient matching and retrieval.  ... 
doi:10.12700/aph.12.1.2015.1.7 fatcat:n2mdfo7ya5asdgilycreceiqdy

SPTS: Single-Point Text Spotting [article]

Dezhi Peng, Xinyu Wang, Yuliang Liu, Jiaxin Zhang, Mingxin Huang, Songxuan Lai, Shenggao Zhu, Jing Li, Dahua Lin, Chunhua Shen, Xiang Bai, Lianwen Jin
2022 arXiv   pre-print
The proposed method is simple yet effective, which can achieve state-of-the-art results on widely used benchmarks.  ...  ., end-to-end text detection and recognition) methods rely on costly bounding box annotations (e.g., text-line, word-level, or character-level bounding boxes).  ...  of the polygonal annotations; the SPTS-Bezier adopts the original metric that matches polygon boxes.  ... 
arXiv:2112.07917v5 fatcat:jggdxgpsn5h3bouen6jtj2eh6a

A shape-based approach to change detection of lakes using time series remote sensing images

Jiang Li, R.M. Narayanan
2003 IEEE Transactions on Geoscience and Remote Sensing  
This paper presents a procedure for efficiently retrieving and representing shapes of interesting features in remotely sensed imagery using supervised classification, object recognition, parametric contour  ...  In addition, shape similarity can be measured by means of a computationally efficient metric.  ...  Mitchell (Cornell University) for providing the source code for the shape similarity measure.  ... 
doi:10.1109/tgrs.2003.817267 fatcat:7fbz4ftapjdytckgrhodkblxcm

Taxonomy of Feature Description Attributes [chapter]

Scott Krig
2014 Computer Vision Metrics  
In addition, existing and novel basis features are being devised in the form of sparse codebooks and visual vocabularies (we use the term basis space loosely). • Polygon Shape Descriptors.  ...  See Chapter 2 for a discussion of wavelets compared to other basis functions. • Morphological shape metrics: predominantly used in the polygon shape descriptor family, composed of shape factors, and referred  ...  Note ■ the fme is intended as a template to capture high-level information for basic analysis.  ... 
doi:10.1007/978-1-4302-5930-5_5 fatcat:kzdhaew74zabbjgsao7s3oycwa

A novel Approach to Human Gait Recognition using possible Speed Invariant features

Anup Nandy, Rupak Chakraborty, Pavan Chakraborty, G.C. Nandi
2014 International Journal of Computational Intelligence Systems  
Initially, all the feature points are represented as the sides of an n-sided polygon for calculating the area of each region.  ...  Classification results demonstrate the ability of proposed feature extraction techniques using Bayes' classifier, distance metrics, and the proposed polynomial based distance metric.  ...  Acknowledgements We owe our sincere thanks and gratitude to Prof Yasushi Yagi and his whole research team for sharing with us the valuable OU-ISIR [30] Gait database without which the work could not  ... 
doi:10.1080/18756891.2014.967004 fatcat:cqfdtyn7arba7apl4szzxps6ga

Scenario Driven In-depth Performance Evaluation of Document Layout Analysis Methods

C. Clausner, S. Pletschacher, A. Antonacopoulos
2011 2011 International Conference on Document Analysis and Recognition  
It combines efficiency and accuracy by using a special interval based geometric representation of regions.  ...  This paper presents an advanced framework for evaluating the performance of layout analysis methods.  ...  Examples are absolute and relative region count deviation (as indicator for over-or under-segmentation) and reading order. The latter is novel in its complexity.  ... 
doi:10.1109/icdar.2011.282 dblp:conf/icdar/ClausnerPA11a fatcat:zhxagyv46rbtfdcqmtmoxvujwa

ProAlignNet: Unsupervised Learning for Progressively Aligning Noisy Contours

VSR Veeravasarapu, Abhishek Goel, Deepak Mittal, Maneesh Singh
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
This work presents a novel ConvNet, "ProAlignNet," that accounts for large scale misalignments and complex transformations between the contour shapes.  ...  similarity metric that uses classical Morphological Chamfer Distance Transform.  ...  Acknowledgments: We thank GEOMNI ( for providing the data for geo-parcel alignment application.  ... 
doi:10.1109/cvpr42600.2020.00969 dblp:conf/cvpr/VeeravasarapuGM20 fatcat:pba5voox2bhfrn6nenwy3xpd5a

DEER: Detection-agnostic End-to-End Recognizer for Scene Text Spotting [article]

Seonghyeon Kim, Seung Shin, Yoonsik Kim, Han-Cheol Cho, Taeho Kil, Jaeheung Surh, Seunghyun Park, Bado Lee, Youngmin Baek
2022 arXiv   pre-print
The proposed method reduces the tight dependency between detection and recognition modules by bridging them with a single reference point for each text instance, instead of using detected regions.  ...  Since only a single point is required to recognize the text, the proposed method enables text spotting without an arbitrarily-shaped detector or bounding polygon annotations.  ...  By relaxing the coupling between the detection of the text bounding polygon and the recognition of text instances, the proposed model allows for the recognition of arbitrary-shaped text instances without  ... 
arXiv:2203.05122v1 fatcat:4atzxtr6gfdxzbzedetttdjpgy

Research on Shape Feature Recognition of B-Rep Model Based on Wavelet Transform

Jihua Wang
2018 Mathematical Problems in Engineering  
, so the shape feature recognition of B-Rep model is worth of much studying.  ...  We present one approach of shape feature recognition of B-Rep model based on the wavelet transform of surface boundary and region; it is inspired by the neuropsychology view that surface is the key visual  ...  Distributed Computer Software Novel Technology.  ... 
doi:10.1155/2018/6310482 fatcat:qcvmkp5nvbgelcl2oveh46yooi

Interest Point Detector and Feature Descriptor Survey [chapter]

Scott Krig
2014 Computer Vision Metrics  
Object Shape Metrics for Blobs and Polygons Object shape metrics are powerful and yield many degrees of freedom with respect to invariance and robustness.  ...  Polygon Shape Descriptors Polygon shape descriptors compute a set of shape features for an arbitrary polygon or blob, and the shape is described using statistical moments or image moments (as discussed  ... 
doi:10.1007/978-1-4302-5930-5_6 fatcat:3jmbq6bpivfqxaq7mcwbs5t2yy

Accurate Polygonal Mapping of Buildings in Satellite Imagery [article]

Bowen Xu, Jiakun Xu, Nan Xue, Gui-Song Xia
2022 arXiv   pre-print
For the Inria dataset, our proposed method also obtains very competitive results on the metrics of IoU and Accuracy. The models and source code are available at  ...  On the AICrowd dataset, our proposed method obtains unanimous improvements on the metrics of AP, APboundary and PoLiS.  ...  The C-IoU metric considers both segmentation accuracy and polygonization complexity.  ... 
arXiv:2208.00609v1 fatcat:7zpzzj3c2rgh3ci3ntixtmkmfu
« Previous Showing results 1 — 15 out of 3,800 results