Filters








26,193 Hits in 7.8 sec

Affine Stable Characteristic based sample expansion for object detection

Ke Gao, Yongdong Zhang, Wei Zhang, Shouxun Lin
2010 Proceedings of the ACM International Conference on Image and Video Retrieval - CIVR '10  
In this paper, we propose a novel method called Affine Stable Characteristic to generate an object feature model using only one object sample.  ...  By integrating affine simulation with stable characteristic mining, a compact and informative object model is generated with high robustness to viewpoint and scale transformations.  ...  AFFINE STABLE CHARACTERISTIC 3.1 System Overview The general approach for our ASC based object detection framework is illustrated in Figure 1 .  ... 
doi:10.1145/1816041.1816103 dblp:conf/civr/GaoZZL10 fatcat:gmreexlz7fccvhdpljsycf6nay

Accurate off-line query expansion for large-scale mobile visual search

Ke Gao, Yongdong Zhang, Dongming Zhang, Shouxun Lin
2013 Signal Processing  
Query expansion is an effective strategy for recall improvement, but existing methods are highly memory and time consuming, and often involve lots of redundant features.  ...  Integrating robust local patch mining and geometric parameter coding, this paper proposes an accurate offline query expansion method for large-scale mobile visual search.  ...  Fig. 4 . 4 Synthetic images automatic generation for each object sample. Fig. 5 . 5 Entropy Loss Ratio (ELR) based robust local patch mining.  ... 
doi:10.1016/j.sigpro.2012.10.011 fatcat:i5hbtpb3rfc4lojzfwhwm42bpq

Visual stem mapping and Geometric Tense coding for Augmented Visual Vocabulary

Ke Gao, Yongdong Zhang, Ping Luo, Wei Zhang, Junhai Xia, Shouxun Lin
2012 2012 IEEE Conference on Computer Vision and Pattern Recognition  
Moreover, the mapping rules can be learned offline with only one sample for each object.  ...  This paper addresses the problem of affine distortions caused by viewpoint changes for the application of image retrieval.  ...  Affine Stable Characteristic (ASC) [5] further selects stable features of ASIFT to reduce its online cost and redundant features.  ... 
doi:10.1109/cvpr.2012.6248059 dblp:conf/cvpr/GaoZLZXL12 fatcat:r6rrufdnkrfl5btjlpbof73ala

Query expansion for VHR image detection

Huaxin Zheng, Huigang Zhang, Xiao Bai, Huijie Zhao
2011 2011 IEEE International Geoscience and Remote Sensing Symposium  
Then we introduce two query expansion baseline expansion and PAS expansion to obtain a new incremental model for re-query.  ...  Inspire by template matching and incremental learning, a new object-oriented object detection methodology for very high resolution remote sensing images is proposed in this paper.  ...  We also plan to test graph based methods for object detection.  ... 
doi:10.1109/igarss.2011.6048928 dblp:conf/igarss/ZhengZBZ11 fatcat:qdxk52rz4rcs7p2q3rga56nsgy

Stereo Correspondence with Local Descriptors for Object Recognition [chapter]

Gee-Sern Jison
2011 Advances in Theory and Applications of Stereo Vision  
have evaluated six affine region detectors, including Harris-affine, Hessian-affine, edge-based region, intensity extrema-based region, salient region and maximally stable extremal region (MSER).  ...  Fig. 1 . 1 Scale invariant interest point detection in affine transformed images: (Top) Initial interest points detected by multi-scale Harris detector with characteristic scales selected by Laplacian  ...  www.intechopen.com/books/advances-in-theory-and-applications-of-stereo-vision/stereocorrespondence-with-local-descriptors-for-object-recognition © 2011 The Author(s).  ... 
doi:10.5772/12886 fatcat:mjcdcj6gijhwvg46bes7ie4nna

Flexible Object Models for Category-Level 3D Object Recognition

Akash Kushal, Cordelia Schmid, Jean Ponce
2007 2007 IEEE Conference on Computer Vision and Pattern Recognition  
Today's category-level object recognition systems largely focus on fronto-parallel views of objects with characteristic texture patterns.  ...  The proposed approach has been implemented, and it outperforms the state-of-the-art algorithms for object detection and localization recently compared in [14] on the Pascal 2005 VOC Challenge Cars Test  ...  These parts are quite discriminative and relatively stable against intra-class variations; however, large viewpoint variations result in the affine model no longer holding for the object parts.  ... 
doi:10.1109/cvpr.2007.383149 dblp:conf/cvpr/KushalSP07 fatcat:blrfkvalr5edhixtbhklanbgtu

Improving SIFT-Based Object Recognition for Robot Applications [chapter]

Patricio Loncomilla, Javier Ruiz-del-Solar
2005 Lecture Notes in Computer Science  
In this article we proposed an improved SIFT-based object recognition methodology for robot applications.  ...  This methodology is employed for implementing a robot-head detection system, which is the main component of a robot gaze direction determination system.  ...  Affine Transformations Verification Based on Geometrical Distortion. A correct detection's affine transformation shouldn't deform very much an object when mapping it.  ... 
doi:10.1007/11553595_133 fatcat:bjpwqbxxhrb27cxke5zuenkhuq

Local Invariant Feature Detectors: A Survey

Tinne Tuytelaars, Krystian Mikolajczyk
2007 Foundations and Trends in Computer Graphics and Vision  
Acknowledgments The authors would like to acknowledge support from the Network of Excellence PASCAL and the Flemish Fund for Scientific Research FWO.  ...  Harris-Affine Given a set of initial points extracted at their characteristic scales based on the Harris-Laplace detection scheme, the iterative estimation of elliptical affine regions as proposed by Lindeberg  ...  Alternatively, a local affine frame can be defined based on a set of stable points along the region contour [135] .  ... 
doi:10.1561/0600000017 fatcat:t5e7h3nhtfgozcn2czsuc52r34

Local Feature Point Extraction Method Based on SIFT

2017 Revista Técnica de la Facultad de Ingeniería Universidad del Zulia  
Based on the feature graph, the object shape recognition algorithm with invariance in translation, scale and rotation can be designed so as to improve the precision and accuracy of the traditional SIFT  ...  It abstracts the edge information of the shape of the image objects into the feature graph constituted by the relative orientations of the feature points and the edges.  ...  Besides, in order to detect the stable key point in the scale space, difference of Gaussian (DOG) is required to construct. For instance, Fig.4 is the scale-space image cluster of DOG.  ... 
doi:10.21311/001.39.11.09 fatcat:vccazolkwbe3fon6ydznpmrwwa

Image Retrieval based on Bag-of-Words model [article]

Jialu Liu
2013 arXiv   pre-print
This article gives a survey for bag-of-words (BoW) or bag-of-features model in image retrieval system.  ...  The goal of this survey is to give an overview of this model and introduce different strategies when building the system based on this model.  ...  We then sequentially discuss the key point detection, local description, vocabulary generation, vector quantization, indexing and search.  ... 
arXiv:1304.5168v1 fatcat:6jycm42bg5guzgzlzdbc2pydby

Second-Order Configuration of Local Features for Geometrically Stable Image Matching and Retrieval

Xiaomeng Wu, Kunio Kashino
2015 IEEE transactions on circuits and systems for video technology (Print)  
We revisit affine shape adaptation and extend its conclusion to characterize the geometrically stable feature of each tuple.  ...  It achieves faster neighborhood association and has a greater robustness to errors in interest point detection and description.  ...  We employ the same detector [30] based on affine shape adaptation [28] , [29] for all datasets to extract interest points.  ... 
doi:10.1109/tcsvt.2014.2382985 fatcat:z473eu53pjbl5oxbkldjtb5ndq

Gaze Direction Determination of Opponents and Teammates in Robot Soccer [chapter]

Patricio Loncomilla, Javier Ruiz-del-Solar
2006 Lecture Notes in Computer Science  
This system is based on a robot-head pose detection system, consisting on two main processing stages: (i) computation of scale-invariant local descriptors of the observed scene, and (ii) matching of these  ...  We aim at reverting this situation by proposing a gaze direction determination system for robot soccer; the system is designed primarily for the four-legged league, but it could be extended to other leagues  ...  Affine Transformations Verification based on Geometrical Distortion. A certain affine transformation shouldn't deform very much an object when mapping it.  ... 
doi:10.1007/11780519_21 fatcat:x5xqlc26grdjlaeackl2q3p54m

Data Association Methods via Video Signal Processing in Imperfect Tracking Scenarios: A Review and Evaluation

Hui Li, Yapeng Liu, Wenzhong Lin, Lingwei Xu, Junyin Wang
2020 Mathematical Problems in Engineering  
Therefore, this article first analyzes characteristics and performance of three traditional data association methods and then focuses on data association methods based on deep learning, which is divided  ...  Finally, it summarizes the current common datasets and evaluation criteria for multiobject tracking and discusses challenges and development trends of data association technology and data association methods  ...  Given a candidate detection and an object tracker sequence as input, SAN extracts combined features by comparing the detection results with each sample in the tracker.  ... 
doi:10.1155/2020/7549816 fatcat:ezj5puaimnbixdvfal2ruvdllm

A multi-view probabilistic model for 3D object classes

Min Sun, Hao Su, S. Savarese, Li Fei-Fei
2009 2009 IEEE Conference on Computer Vision and Pattern Recognition  
Our approach is among the first to propose a generative probabilistic framework for 3D object categorization.  ...  We propose a novel probabilistic framework for learning visual models of 3D object categories by combining appearance information and geometric constraints.  ...  Acknowledgments The authors would like to thank Jia Deng, Barry Chai, and Bangpeng Yao for their help for data collection.  ... 
doi:10.1109/cvprw.2009.5206723 fatcat:6lujcm42mnfcrm45vbv3xcllmi

A multi-view probabilistic model for 3D object classes

Min Sun, Hao Su, Silvio Savarese, Li Fei-Fei
2009 2009 IEEE Conference on Computer Vision and Pattern Recognition  
Our approach is among the first to propose a generative probabilistic framework for 3D object categorization.  ...  We propose a novel probabilistic framework for learning visual models of 3D object categories by combining appearance information and geometric constraints.  ...  Acknowledgments The authors would like to thank Jia Deng, Barry Chai, and Bangpeng Yao for their help for data collection.  ... 
doi:10.1109/cvpr.2009.5206723 dblp:conf/cvpr/SunSS009 fatcat:fumfglyfybg7hdjkbmz5fsc4ty
« Previous Showing results 1 — 15 out of 26,193 results