Filters








20,898 Hits in 6.6 sec

Multi-stage Contour Based Detection of Deformable Objects [chapter]

Saiprasad Ravishankar, Arpit Jain, Anurag Mittal
2008 Lecture Notes in Computer Science  
We present an efficient multi stage approach to detection of deformable objects in real, cluttered images given a single or few hand drawn examples as models.  ...  In the fine matching stage, the entire object contour in the localized regions is built from the k-segment groups and given a comprehensive score in a method that uses dynamic programming.  ...  Object segments in the image which could still not be correctly matched at this stage due to their larger deformation are efficiently detected in our final detection stage.  ... 
doi:10.1007/978-3-540-88682-2_37 fatcat:mhoq4lupb5bxbixail7gbljfly

Deep Snake for Real-Time Instance Segmentation

Sida Peng, Wen Jiang, Huaijin Pi, Xiuli Li, Hujun Bao, Xiaowei Zhou
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Based on deep snake, we develop a two-stage pipeline for instance segmentation: initial contour proposal and contour deformation, which can handle errors in object localization.  ...  Unlike some recent methods that directly regress the coordinates of the object boundary points from an image, deep snake uses a neural network to iteratively deform an initial contour to match the object  ...  Based on deep snake, we developed a two-stage pipeline for instance segmentation: initial contour proposal and contour deformation.  ... 
doi:10.1109/cvpr42600.2020.00856 dblp:conf/cvpr/PengJPLBZ20 fatcat:ubz3m3a3rfcpjlbuzn4mticnl4

Deep Snake for Real-Time Instance Segmentation [article]

Sida Peng, Wen Jiang, Huaijin Pi, Xiuli Li, Hujun Bao, Xiaowei Zhou
2020 arXiv   pre-print
Based on deep snake, we develop a two-stage pipeline for instance segmentation: initial contour proposal and contour deformation, which can handle errors in object localization.  ...  Unlike some recent methods that directly regress the coordinates of the object boundary points from an image, deep snake uses a neural network to iteratively deform an initial contour to match the object  ...  Based on deep snake, we developed a two-stage pipeline for instance segmentation: initial contour proposal and contour deformation.  ... 
arXiv:2001.01629v3 fatcat:oagj2gypxzar5hezqsgicdjs4e

Automatic Aortic Root Segmentation with Shape Constraints and Mesh Regularisation

Robert Ieuan Palmer, Xianghua Xie, Gary Tam
2015 Procedings of the British Machine Vision Conference 2015  
The deformable segmentation consists of a learning-based boundary detector, where each vertex of the mesh template is adjusted to fit the object boundary along the orthogonal direction.  ...  The estimation of the aortic root pose can be considered as an object detection problem, and a marginal space learning (MSL) method is adopted from [3] for this purpose.  ...  The deformable segmentation consists of a learning-based boundary detector, where each vertex of the mesh template is adjusted to fit the object boundary along the orthogonal direction.  ... 
doi:10.5244/c.29.83 dblp:conf/bmvc/PalmerXT15 fatcat:l4dqrxx2hfcgrdeofrgqi3v5cy

Energy matching based on deformable templates

Y.S. Tawfik, A.M. Darwish, S.I. Shaheen
1999 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258)  
The approach is based on deformable templates and was tested and proven be very robust with respect to changes in object scale, position and orientation as well as noise and local deformations of shape  ...  In this paper, we present a scheme for object localization and classification based on inexact description of the shape.  ...  (c) Detected objects. Detecting multiple objects with different scale, orientation, and local deformation. (a) Template. (b) Input image. (c) Detected objects. Classifying objects. (a) Input image.  ... 
doi:10.1109/icassp.1999.757592 dblp:conf/icassp/TawfikDS99 fatcat:ckrryrsggngpffdghyzrc2pqti

An Automated Multimodal Spectral Cluster Based Segmentation for Tumor and Lesion Detection in Pet Images

M. Manoj, L. Padmasuresh
2016 Research Journal of Applied Sciences Engineering and Technology  
A Multi-Spectral Contour Cluster (MSCC) mechanism is introduced that organizes the spectral contour features of shape, size, location and intensity into multi-spectral clusters for quicker segmentation  ...  The Spectral Contours with Constrained Threshold (SCCT) technique in AMSCS is carried out to various spectral features of the PET image without any deformation, improving the true positive rate.  ...  Multi-Spectral Contour Cluster (MSCC) mechanism: The second step involved in the design of Automated Multimodal Spectral Cluster based Segmentation is the construction of Multi-Spectral Contour Cluster  ... 
doi:10.19026/rjaset.12.2679 fatcat:4rs5sudb7vcdnl7fca5qzdn54m

Shape-Regulated Particle Filtering for Tracking Non-Rigid Objects

Jie Shao, Rama Chellappa, Fatih Porikli
2006 2006 International Conference on Image Processing  
Second, by using a dynamic object model, we generate a probabilistic map of deformation to reshape its contour.  ...  This paper presents an active contour based algorithm for tracking non-rigid objects in heavily cluttered scenes.  ...  Contour-based tracking approaches attract lots of interest due to their ability to accurately delineate contours or boundaries of objects.  ... 
doi:10.1109/icip.2006.312993 dblp:conf/icip/ShaoCP06 fatcat:xpqqi5te4rc3rh23fcdctagiv4

Automatic 3D Car Model Alignment for Mixed Image-Based Rendering

Rodrigo Ortiz-Cayon, Abdelaziz Djelouah, Francisco Massa, Mathieu Aubry, George Drettakis
2016 2016 Fourth International Conference on 3D Vision (3DV)  
We do this by first adapting learning-based methods to detect and identify an object class and pose in images.  ...  These steps provide models which are well-aligned in 3D and to contours in all the images of the multi-view dataset, allowing us to use the resulting model in our mixed IBR algorithm.  ...  Multi-View Object Class Detection We use the multi-region R-CNN [16] -one of the topperforming detection algorithms.  ... 
doi:10.1109/3dv.2016.37 dblp:conf/3dim/CayonDMAD16 fatcat:leivwqba2fdanar3ahlfhsqjhq

Multiple-Cue-Based Visual Object Contour Tracking with Incremental Learning [chapter]

Aiping Wang, Zhi-Quan Cheng, Ralph R. Martin, Sikun Li
2013 Lecture Notes in Computer Science  
This paper proposes a visual object contour tracking algorithm using a multi-cue fusion particle filter.  ...  Both of the above are taken as observation models for contour deformation; these models are fused efficiently using a mixed cascade sampling process.  ...  In this paper, we propose a multi-cue based discriminative object contour tracking algorithm with two major contributions as follows. 1.  ... 
doi:10.1007/978-3-642-37042-7_16 fatcat:ryxiru7oqrbifjbrhkbv7gcin4

A learning algorithm for model based object detection

Chen Guodong, Zeyang Xia, Rongchuan Sun, Zhenhua Wang, Zhiwu Ren, Lining Sun
2011 2011 8th International Conference on Ubiquitous Robots and Ambient Intelligence (URAI)  
Findings -The major contributions of this paper are the application of learned shape fragments-based model for object detection in complex environment and a novel two-stage object detection framework.  ...  The purpose of this paper is to introduce a model-based object detection method which uses only shape-fragment features.  ...  Most of the approaches employ locally deformable 2D models for discrete viewpoints. This casts the 3D object detection problem into 2D object detection in multi-views.  ... 
doi:10.1109/urai.2011.6145941 dblp:conf/urai/ChenXSWRS11 fatcat:b6ng2a5srrbbxdgojquqxnk53q

E2EC: An End-to-End Contour-based Method for High-Quality High-Speed Instance Segmentation [article]

Tao Zhang, Shiqing Wei, Shunping Ji
2022 arXiv   pre-print
This consists of a contour initialization module for constructing more explicit learning goals and a global contour deformation module for taking advantage of all of the vertices' features better.  ...  In this paper, we introduce a novel contour-based method, named E2EC, for high-quality instance segmentation.  ...  The multi-stage methods iteratively adjust the initial contour based on the features of the contour vertices to obtain a more refined segmentation result.  ... 
arXiv:2203.04074v1 fatcat:qp756pgebbhedngu4cizc5xjkm

Real Time Object Detection & Tracking over a Mobile Platform

M. A. Mekhtiche, M. A. Bencherif, M. Algabri, M. Alsulaiman, R. Hedjar, M. Faisal, K. AlMutib
2016 Indian Journal of Science and Technology  
The method is based on a new vectored contour to identify objects from HSV images.  ...  The aim of the proposed method is to let the robot identify multi-targets within a scene, then move toward the desired object.  ...  Keywords: Mobile Robot, Object Recognition and Tracking The suggested normalized vectored contour is then fed as input to the multi-object identification stage, as illustrated in Figure 1.  ... 
doi:10.17485/ijst/2015/v8i32/92131 fatcat:bkxdqko7lrazljfnyfhdmtcrjy

FisheyeDet: A Self-study and Contour-based Object Detector in Fisheye Images

Tangwei Li, Guanjun Tong, Hongying Tang, Baoqing Li, Bo Chen
2020 IEEE Access  
contour of distorted objects as the core.  ...  Finally, due to the lack of public fisheye datasets, we are on the first attempt to create a multi-class fisheye dataset VOC-Fisheye for object detection.  ...  correction-based methods usually include two stages: image warping and object detection.  ... 
doi:10.1109/access.2020.2987868 fatcat:hbbuvheifjb2nc4fc3q6uxoc2e

An Intelligent Fire-Detection Method Based on Image Processing

Ming Xiu Lin, Wen Lin Chen, Bao Song Liu, Li Na Hao
2011 Advanced Engineering Forum  
To achieve fully automatic surveillance of fires, an intelligent fire detection method based on a multi-stage decision strategy of image processing is proposed.  ...  Multi-stage decision strategy can improve the algorithm performance and reduce false-alarm rate.  ...  In this paper, to achieve fully automatic surveillance of fires, an intelligent fire detection method based on a multi-stage decision strategy of image processing is proposed.  ... 
doi:10.4028/www.scientific.net/aef.2-3.172 fatcat:mdmrwfqrgrelpfciueyhsxunqe

Modeling from contour drawings

Vladislav Kraevoy, Alla Sheffer, Michiel van de Panne
2009 Proceedings of the 6th Eurographics Symposium on Sketch-Based Interfaces and Modeling - SBIM '09  
We investigate the development of 3D models from multi-stroke contour drawings with the help of a 3D template model that serves as a shape prior.  ...  Occlusion contours are a natural feature to draw when tracing an object in an image or when drawing an object.  ...  We explore the alternative option of directly interpreting a multi-stroke contour drawing.  ... 
doi:10.1145/1572741.1572749 fatcat:ztkwnyaqkbb6xguuwfkijmohu4
« Previous Showing results 1 — 15 out of 20,898 results