The Experts below are selected from a list of 46059 Experts worldwide ranked by ideXlab platform
Richard Szeliski - One of the best experts on this subject based on the ideXlab platform.
-
image Alignment and stitching
Handbook of Mathematical Models in Computer Vision, 2006Co-Authors: Richard SzeliskiAbstract:Stitching multiple images together to create beautiful high-resolution panoramas is one of the most popular consumer applications of image registration and blending. In this chapter, I review the motion models (geometric transformations) that underlie panoramic image stitching, discuss direct intensity-based and feature-based registration algorithms, and present global and Local Alignment techniques needed to establish high-accuracy correspondences between overlapping images. I then discuss various compositing options, including multi band and gradient-domain blending, as well as techniques for removing blur and ghosted images. The resulting techniques can be used to create high-quality panoramas for static or interactive viewing.
-
construction of panoramic image mosaics with global and Local Alignment
Panoramic vision, 2001Co-Authors: Heung-yeung Shum, Richard SzeliskiAbstract:The automatic construction of large, high-resolution image mosaics is an active area of research in the fields of photogrammetry, computer vision, image processing, and computer graphics. Image mosaics can be used for many different applications [163, 1.22]. The most traditional application is the construction of large aerial and satellite photographs from collections of images [186]. More recent applications include scene stabilization and change detection [93], video compression [125, 122, 167] and video indexing [240], increasing the field of view [105, 177, 266] and resolution [126, 50] of a camera, and even simple photo editing [38]. A particularly popular application is the emulation of traditional film-based panoramic photography [175] with digital panoramic mosaics, for applications such as the construction of virtual environments [181, 267] and virtual travel [49].
-
Systems and experiment paper: Construction of panoramic image mosaics with global and Local Alignment
International Journal of Computer Vision, 2000Co-Authors: Heung-yeung Shum, Richard SzeliskiAbstract:This paper presents a complete system for constructing panoramic image mosaics from sequences of images. Our mosaic representation also associates a transformation matrix with each input image, rather than explicitly projecting all of the images onto a common surface (e.g., a cylinder). In particular, to construct a full view panorama, we introduce a rotational mosaic representation that associates a rotation matrix (and optionally a focal length) with each input image. A patch-based Alignment algorithm is developed to quickly align two images given motion models. Techniques for estimating and refining camera focal lengths are also presented.\n\nIn order to reduce accumulated registration errors, we apply global Alignment (block adjustment) to the whole sequence of images, which results in an optimally registered image mosaic. To compensate for small amounts of motion parallax introduced by translations of the camera and other unmodeled distortions, we use a Local Alignment (deghosting) technique which warps each image based on the results of pairwise Local image registrations. By combining both global and Local Alignment, we significantly improve the quality of our image mosaics, thereby enabling the creation of full view panoramic mosaics with hand-held cameras.\n\nWe also present an inverse texture mapping algorithm for efficiently extracting environment maps from our panoramic image mosaics. By mapping the mosaic onto an arbitrary texture-mapped polyhedron surrounding the origin, we can explore the virtual environment using standard 3D graphics viewers and hardware without requiring special-purpose players.
-
construction and refinement of panoramic mosaics with global and Local Alignment
International Conference on Computer Vision, 1998Co-Authors: Heung-yeung Shum, Richard SzeliskiAbstract:This paper presents techniques for constructing full view panoramic mosaics form sequences of images. Our representation associates a rotation matrix (and optionally a focal length) with each input image, rather than explicitly projecting all of the images onto a common surface (e.g., a cylinder). In order to reduce accumulated registration errors we apply global Alignment (block adjustment) to whole sequence of images, which results in an optimal image mosaic (in the least squares sense). To compensate for small amounts of motion parallax introduced by translations of the camera and other unmodeled distortions we develop a Local Alignment (deghosting) technique which warps each image based on the results of pairwise Local image registrations. By combining both global and Local Alignment we significantly improve the quality of our image mosaics thereby enabling the creation of full view panoramic mosaics with hand-held cameras.
Heung-yeung Shum - One of the best experts on this subject based on the ideXlab platform.
-
construction of panoramic image mosaics with global and Local Alignment
Panoramic vision, 2001Co-Authors: Heung-yeung Shum, Richard SzeliskiAbstract:The automatic construction of large, high-resolution image mosaics is an active area of research in the fields of photogrammetry, computer vision, image processing, and computer graphics. Image mosaics can be used for many different applications [163, 1.22]. The most traditional application is the construction of large aerial and satellite photographs from collections of images [186]. More recent applications include scene stabilization and change detection [93], video compression [125, 122, 167] and video indexing [240], increasing the field of view [105, 177, 266] and resolution [126, 50] of a camera, and even simple photo editing [38]. A particularly popular application is the emulation of traditional film-based panoramic photography [175] with digital panoramic mosaics, for applications such as the construction of virtual environments [181, 267] and virtual travel [49].
-
Systems and experiment paper: Construction of panoramic image mosaics with global and Local Alignment
International Journal of Computer Vision, 2000Co-Authors: Heung-yeung Shum, Richard SzeliskiAbstract:This paper presents a complete system for constructing panoramic image mosaics from sequences of images. Our mosaic representation also associates a transformation matrix with each input image, rather than explicitly projecting all of the images onto a common surface (e.g., a cylinder). In particular, to construct a full view panorama, we introduce a rotational mosaic representation that associates a rotation matrix (and optionally a focal length) with each input image. A patch-based Alignment algorithm is developed to quickly align two images given motion models. Techniques for estimating and refining camera focal lengths are also presented.\n\nIn order to reduce accumulated registration errors, we apply global Alignment (block adjustment) to the whole sequence of images, which results in an optimally registered image mosaic. To compensate for small amounts of motion parallax introduced by translations of the camera and other unmodeled distortions, we use a Local Alignment (deghosting) technique which warps each image based on the results of pairwise Local image registrations. By combining both global and Local Alignment, we significantly improve the quality of our image mosaics, thereby enabling the creation of full view panoramic mosaics with hand-held cameras.\n\nWe also present an inverse texture mapping algorithm for efficiently extracting environment maps from our panoramic image mosaics. By mapping the mosaic onto an arbitrary texture-mapped polyhedron surrounding the origin, we can explore the virtual environment using standard 3D graphics viewers and hardware without requiring special-purpose players.
-
construction and refinement of panoramic mosaics with global and Local Alignment
International Conference on Computer Vision, 1998Co-Authors: Heung-yeung Shum, Richard SzeliskiAbstract:This paper presents techniques for constructing full view panoramic mosaics form sequences of images. Our representation associates a rotation matrix (and optionally a focal length) with each input image, rather than explicitly projecting all of the images onto a common surface (e.g., a cylinder). In order to reduce accumulated registration errors we apply global Alignment (block adjustment) to whole sequence of images, which results in an optimal image mosaic (in the least squares sense). To compensate for small amounts of motion parallax introduced by translations of the camera and other unmodeled distortions we develop a Local Alignment (deghosting) technique which warps each image based on the results of pairwise Local image registrations. By combining both global and Local Alignment we significantly improve the quality of our image mosaics thereby enabling the creation of full view panoramic mosaics with hand-held cameras.
Qi Tian - One of the best experts on this subject based on the ideXlab platform.
-
glad global Local Alignment descriptor for scalable person re identification
IEEE Transactions on Multimedia, 2019Co-Authors: Longhui Wei, Shiliang Zhang, Hantao Yao, Wen Gao, Qi TianAbstract:The huge variance of human pose and the misalign-ment of detected human images significantly increase the difficulty of pedestrian image matching in person Re-Identification (Re-ID). Moreover, the massive visual data being produced by surveillance video cameras requires highly efficient person Re-ID systems. Targeting to solve the first problem, this work proposes a robust and discriminative pedestrian image descriptor, namely, the Global–Local-Alignment Descriptor (GLAD). For the second problem, this work treats person Re-ID as image retrieval and proposes an efficient indexing and retrieval framework. GLAD explicitly leverages the Local and global cues in the human body to generate a discriminative and robust representation. It consists of part extraction and descriptor learning modules, where several part regions are first detected and then deep neural networks are designed for representation learning on both the Local and global regions. A hierarchical indexing and retrieval framework is designed to perform offline relevance mining to eliminate the huge person ID redundancy in the gallery set, and accelerate the online Re-ID procedure. Extensive experimental results on widely used public benchmark datasets show GLAD achieves competitive accuracy compared to the state-of-the-art methods. On a large-scale person, with the Re-ID dataset containing more than 520 K images, our retrieval framework significantly accelerates the online Re-ID procedure while also improving Re-ID accuracy. Therefore, this work has the potential to work better on person Re-ID tasks in real scenarios.
-
glad global Local Alignment descriptor for pedestrian retrieval
arXiv: Computer Vision and Pattern Recognition, 2017Co-Authors: Longhui Wei, Shiliang Zhang, Hantao Yao, Wen Gao, Qi TianAbstract:The huge variance of human pose and the misAlignment of detected human images significantly increase the difficulty of person Re-Identification (Re-ID). Moreover, efficient Re-ID systems are required to cope with the massive visual data being produced by video surveillance systems. Targeting to solve these problems, this work proposes a Global-Local-Alignment Descriptor (GLAD) and an efficient indexing and retrieval framework, respectively. GLAD explicitly leverages the Local and global cues in human body to generate a discriminative and robust representation. It consists of part extraction and descriptor learning modules, where several part regions are first detected and then deep neural networks are designed for representation learning on both the Local and global regions. A hierarchical indexing and retrieval framework is designed to eliminate the huge redundancy in the gallery set, and accelerate the online Re-ID procedure. Extensive experimental results show GLAD achieves competitive accuracy compared to the state-of-the-art methods. Our retrieval framework significantly accelerates the online Re-ID procedure without loss of accuracy. Therefore, this work has potential to work better on person Re-ID tasks in real scenarios.
Moonjin Kang - One of the best experts on this subject based on the ideXlab platform.
-
a rigorous derivation from the kinetic cucker smale model to the pressureless euler system with nonLocal Alignment
arXiv: Analysis of PDEs, 2017Co-Authors: Alessio Figalli, Moonjin KangAbstract:We consider the kinetic Cucker-Smale model with Local Alignment as a mesoscopic description for the flocking dynamics. The Local Alignment was first proposed by Karper, Mellet and Trivisa \cite{K-M-T-3}, as a singular limit of a normalized non-symmetric Alignment introduced by Motsch and Tadmor \cite{M-T-1}. The existence of weak solutions to this model is obtained in \cite{K-M-T-3}. The time-asymptotic flocking behavior is shown in this article. Our main contribution is to provide a rigorous derivation from mesoscopic to mascroscopic description for the Cucker-Smale flocking models. More precisely, we prove the hydrodynamic limit of the kinetic Cucker-Smale model with Local Alignment towards the pressureless Euler system with nonLocal Alignment, under a regime of strong Local Alignment. Based on the relative entropy method, a main difficulty in our analysis comes from the fact that the entropy of the limit system has no strictly convexity in terms of density variable. To overcome this, we combine relative entropy quantities with the 2-Wasserstein distance.
-
asymptotic analysis of vlasov type equations under strong Local Alignment regime
Mathematical Models and Methods in Applied Sciences, 2015Co-Authors: Moonjin Kang, Alexis F VasseurAbstract:We consider the hydrodynamic limit of a collisionless and non-diffusive kinetic equation under strong Local Alignment regime. The Local Alignment is first considered by Karper, Mellet and Trivisa in [On strong Local Alignment in the kinetic Cucker–Smale model, in Hyperbolic Conservation Laws and Related Analysis with Applications, Springer Proceedings in Mathematics & Statistics, Vol. 49 (Springer, 2014), pp. 227–242], as a singular limit of an Alignment force proposed by Motsch and Tadmor in [A new model for self-organized dynamics and its flocking behavior, J. Statist. Phys. 141 (2011) 923–947]. As the Local Alignment strongly dominates, a weak solution to the kinetic equation under consideration converges to the Local equilibrium, which has the form of mono-kinetic distribution. We use the relative entropy method and weak compactness to rigorously justify the weak convergence of our kinetic equation to the pressureless Euler system.
-
asymptotic analysis of vlasov type equations under strong Local Alignment regime
arXiv: Analysis of PDEs, 2014Co-Authors: Moonjin Kang, Alexis F VasseurAbstract:We consider the hydrodynamic limit of a collisionless and non-diffusive kinetic equation under strong Local Alignment regime. The Local Alignment is first considered by Karper, Mellet and Trivisa in [24], as a singular limit of an Alignment force proposed by Motsch and Tadmor in [32]. As the Local Alignment strongly dominate, a weak solution to the kinetic equation under consideration converges to the Local equilibrium, which has the form of mono-kinetic distribution. We use the relative entropy method and weak compactness to rigorously justify the weak convergence of our kinetic equation to the pressureless Euler system.
Thomas L Madden - One of the best experts on this subject based on the ideXlab platform.
-
new finite size correction for Local Alignment score distributions
BMC Research Notes, 2012Co-Authors: Yonil Park, Sergey L Sheetlin, Thomas L Madden, John L SpougeAbstract:Background Local Alignment programs often calculate the probability that a match occurred by chance. The calculation of this probability may require a “finite-size” correction to the lengths of the sequences, as an Alignment that starts near the end of either sequence may run out of sequence before achieving a significant score.
-
NCBI BLAST: a better web interface
Nucleic acids research, 2008Co-Authors: Mark R. Johnson, Irena Zaretskaya, Yan Raytselis, Yuri Merezhuk, Scott D. Mcginnis, Thomas L MaddenAbstract:Basic Local Alignment Search Tool (BLAST) is a sequence similarity search program. The public interface of BLAST, http://www.ncbi.nlm.nih.gov/ blast, at the NCBI website has recently been reengineered to improve usability and performance. Key new features include simplified search forms, improved navigation, a list of recent BLAST results, saved search strategies and a documentation directory. Here, we describe the BLAST web application’s new features, explain design decisions and outline plans for future improvement.
-
blast at the core of a powerful and diverse set of sequence analysis tools
Nucleic Acids Research, 2004Co-Authors: Scott D. Mcginnis, Thomas L MaddenAbstract:Basic Local Alignment Search Tool (BLAST) is one of the most heavily used sequence analysis tools available in the public domain. There is now a wide choice of BLAST algorithmsthatcanbeusedtosearch many different sequence databases via the BLAST web pages (http://www.ncbi.nlm.nih.gov/BLAST/). All the algorithm–database combinations can be executed with default parameters or with customized settings, and the results can be viewed in a variety of ways. A new online resource, the BLAST Program Selection Guide, has been created to assist in the definition of search strategies. This article discusses optimal search strategies and highlights some BLAST features that can make your searches more powerful.
