The Experts below are selected from a list of 360 Experts worldwide ranked by ideXlab platform
M Wilkinson - One of the best experts on this subject based on the ideXlab platform.
-
an efficient Attribute Space connected filter on graphs to reconstruct paths in point clouds
Pattern Recognition, 2020Co-Authors: M Babai, N Kalantarnayestanaki, Johan G Messchendorp, M WilkinsonAbstract:Abstract Measurements by many multi-sensor systems can be considered as point-clouds. One such system is the tracker for the PANDA experiment. Charged particles passing through the tracker produce patterns representing their paths. We present a new, graph-based, Attribute-Space morphological connected filter for reconstructing particle paths through such a detector. We introduce the concept of Attribute-Spaces and Attribute-Space connected filters on graphs, rather than binary images and show a new processing scheme to reduce the size of the memory required to store the Attribute-Space representations of binary images and graphs. The result is an O(Nlog (N)) algorithm with a total recognition error of approximately 0.10, a significant improvement compared to our previous state-of-the-art O(N2) algorithm with a total error of 0.17.
-
tracking sub atomic particles through the Attribute Space
International Symposium on Memory Management, 2016Co-Authors: M Babai, N Kalantarnayestanaki, J G Messchendorp, M WilkinsonAbstract:In this paper, we present the results of an application of Attribute Space morphological filters for tracking sub-atomic particles in magnetic fields. For this purpose, we have applied the concept of Attribute Space and connectivity to the binary images produced by charged particles passing through the tracking detector for the future experiment PANDA. This detector could be considered as an undirected graph with irregular neighbourhood relations. For this project, we rely only on the detector geometry. In addition, we have extended the graph to estimate the z-coordinates of the particle paths. The result is an O(n 2), proof of concept algorithm with a total error of approximately 0.17. The results look promising; however, more work needs to be done to make this algorithm applicable for the real-life case.
-
hyperconnectivity Attribute Space connectivity and path openings theoretical relationships
International Symposium on Mathematical Morphology and Its Application to Signal and Image Processing, 2009Co-Authors: M WilkinsonAbstract:In this paper the relationship of hyperconnected filters with path openings and Attribute-Space connected filters is studied. Using a recently developed axiomatic framework based on hyperconnectivity operators, which are the hyperconnected equivalents of connectivity openings, it is shown that path openings are a special case of hyperconnected area openings. The new axiomatics also yield insight into the relationship between hyperconnectivity and Attribute-Space connectivity. It is shown any hyperconnectivity is an Attribute-Space connectivity, but that the reverse is not true.
-
Attribute Space connectivity and connected filters
Image and Vision Computing, 2007Co-Authors: M WilkinsonAbstract:In this paper connected operators from mathematical morphology are extended to a wider class of operators, which are based on connectivities in higher dimensional Spaces, similar to scale Spaces, which will be called Attribute-Spaces. Though some properties of connected filters are lost, granulometries can be defined under certain conditions, and pattern spectra in most cases. The advantage of this approach is that regions can be split into constituent parts before filtering more naturally than by using partitioning connectivities. Furthermore, the approach allows dealing with overlap, which is impossible in connectivity. A theoretical comparison to hyperconnectivity suggests the new concept is different. The theoretical results are illustrated by several examples. These show how Attribute-Space connected filters merge the ability of filtering based on local structure using classical, structuring-element-based filters to the object-Attribute-based filtering of connected filters, and how this differs from similar attempts using second-generation connectivity.
-
Attribute Space connected filters
International Symposium on Memory Management, 2005Co-Authors: M WilkinsonAbstract:In this paper connected operators from mathematical morphology are extended to a wider class of operators, which are based on connectivities in higher dimension Spaces, similar to scale Spaces which will be called Attribute Spaces. Though some properties of connected filters are lost, granulometries can be defined under certain conditions, and pattern spectra in most cases. The advantage of this approach is that regions can be split into constituent parts before filtering more naturally than by using partitioning connectivities.
M Babai - One of the best experts on this subject based on the ideXlab platform.
-
an efficient Attribute Space connected filter on graphs to reconstruct paths in point clouds
Pattern Recognition, 2020Co-Authors: M Babai, N Kalantarnayestanaki, Johan G Messchendorp, M WilkinsonAbstract:Abstract Measurements by many multi-sensor systems can be considered as point-clouds. One such system is the tracker for the PANDA experiment. Charged particles passing through the tracker produce patterns representing their paths. We present a new, graph-based, Attribute-Space morphological connected filter for reconstructing particle paths through such a detector. We introduce the concept of Attribute-Spaces and Attribute-Space connected filters on graphs, rather than binary images and show a new processing scheme to reduce the size of the memory required to store the Attribute-Space representations of binary images and graphs. The result is an O(Nlog (N)) algorithm with a total recognition error of approximately 0.10, a significant improvement compared to our previous state-of-the-art O(N2) algorithm with a total error of 0.17.
-
a graph formalism for time and memory efficient morphological Attribute Space connected filters
International Symposium on Memory Management, 2019Co-Authors: M Babai, Ananda S Chowdhury, Michael H F WilkinsonAbstract:Attribute-Space connectivity has been put forward as a means of improving image segmentation in the case of overlapping structures. Its main drawback is the huge memory load incurred by mapping a N-dimensional image to an \((N + \dim (A))\)-dimensional volume, with \(\dim (A)\) the dimensionality of the Attribute vectors used. In this theoretical paper we introduce a more Space and time efficient scheme, by representing Attribute Spaces for analysis of binary images as a graph rather than a volume. Introducing a graph formalism for Attribute-Space connectivity opens up the possibility of using Attribute-Space connectivity on 3D volumes or using more than one Attribute dimension, without incurring huge memory costs. Furthermore, the graph formalism does not require quantization of the Attribute values, as is the case when representing Attribute Spaces in terms of \((N + \dim (A))\)-dimensional discrete volumes. Efficient processing of high dimensional data produced by multi-sensor detection systems is another advantage of application of our formalism.
-
tracking sub atomic particles through the Attribute Space
International Symposium on Memory Management, 2016Co-Authors: M Babai, N Kalantarnayestanaki, J G Messchendorp, M WilkinsonAbstract:In this paper, we present the results of an application of Attribute Space morphological filters for tracking sub-atomic particles in magnetic fields. For this purpose, we have applied the concept of Attribute Space and connectivity to the binary images produced by charged particles passing through the tracking detector for the future experiment PANDA. This detector could be considered as an undirected graph with irregular neighbourhood relations. For this project, we rely only on the detector geometry. In addition, we have extended the graph to estimate the z-coordinates of the particle paths. The result is an O(n 2), proof of concept algorithm with a total error of approximately 0.17. The results look promising; however, more work needs to be done to make this algorithm applicable for the real-life case.
Arie E Kaufman - One of the best experts on this subject based on the ideXlab platform.
-
modified dendrogram of Attribute Space for multidimensional transfer function design
IEEE Transactions on Visualization and Computer Graphics, 2012Co-Authors: Lei Wang, Xin Zhao, Arie E KaufmanAbstract:We introduce a modified dendrogram (MD) (with subtrees to represent clusters) and display it in 2D for multidimensional transfer function design. Such a transfer function for direct volume rendering employs a multidimensional Space, termed Attribute Space. The MD reveals the hierarchical structure information of the Attribute Space. The user can design a transfer function in an intuitive and informative manner using the MD user interface in 2D instead of multidimensional Space, where it is hard to ascertain the relationship of the Space. In addition, we provide the capability to interactively modify the granularity of the MD. The coarse-grained MD primarily shows the global information of the Attribute Space while the fine-grained MD reveals the finer details, and the separation ability of the Attribute Space is completely preserved in the finest granularity. With this so called multigrained method, the user can efficiently create a transfer function using the coarse-grained MD, and then fine tune it with the fine-grained MDs. Our method is independent on the type of the Attributes and supports arbitrary-dimension Attribute Space.
Michael H F Wilkinson - One of the best experts on this subject based on the ideXlab platform.
-
a graph formalism for time and memory efficient morphological Attribute Space connected filters
International Symposium on Memory Management, 2019Co-Authors: M Babai, Ananda S Chowdhury, Michael H F WilkinsonAbstract:Attribute-Space connectivity has been put forward as a means of improving image segmentation in the case of overlapping structures. Its main drawback is the huge memory load incurred by mapping a N-dimensional image to an \((N + \dim (A))\)-dimensional volume, with \(\dim (A)\) the dimensionality of the Attribute vectors used. In this theoretical paper we introduce a more Space and time efficient scheme, by representing Attribute Spaces for analysis of binary images as a graph rather than a volume. Introducing a graph formalism for Attribute-Space connectivity opens up the possibility of using Attribute-Space connectivity on 3D volumes or using more than one Attribute dimension, without incurring huge memory costs. Furthermore, the graph formalism does not require quantization of the Attribute values, as is the case when representing Attribute Spaces in terms of \((N + \dim (A))\)-dimensional discrete volumes. Efficient processing of high dimensional data produced by multi-sensor detection systems is another advantage of application of our formalism.
Shaogang Gong - One of the best experts on this subject based on the ideXlab platform.
-
Semantic Autoencoder for Zero-Shot Learning
2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017Co-Authors: Elyor Kodirov, Tao Xiang, Shaogang GongAbstract:Existing zero-shot learning (ZSL) models typically learn a projection function from a feature Space to a semantic embedding Space (e.g. Attribute Space). However, such a projection function is only concerned with predicting the training seen class semantic representation (e.g. Attribute prediction) or classification. When applied to test data, which in the context of ZSL contains different (unseen) classes without training data, a ZSL model typically suffers from the project domain shift problem. In this work, we present a novel solution to ZSL based on learning a Semantic AutoEncoder (SAE). Taking the encoder-decoder paradigm, an encoder aims to project a visual feature vector into the semantic Space as in the existing ZSL models. However, the decoder exerts an additional constraint, that is, the projection/code must be able to reconstruct the original visual feature. We show that with this additional reconstruction constraint, the learned projection function from the seen classes is able to generalise better to the new unseen classes. Importantly, the encoder and decoder are linear and symmetric which enable us to develop an extremely efficient learning algorithm. Extensive experiments on six benchmark datasets demonstrate that the proposed SAE outperforms significantly the existing ZSL models with the additional benefit of lower computational cost. Furthermore, when the SAE is applied to supervised clustering problem, it also beats the state-of-the-art.
-
zero shot object recognition by semantic manifold distance
Computer Vision and Pattern Recognition, 2015Co-Authors: Tao Xiang, Elyor Kodirov, Shaogang GongAbstract:Object recognition by zero-shot learning (ZSL) aims to recognise objects without seeing any visual examples by learning knowledge transfer between seen and unseen object classes. This is typically achieved by exploring a semantic embedding Space such as Attribute Space or semantic word vector Space. In such a Space, both seen and unseen class labels, as well as image features can be embedded (projected), and the similarity between them can thus be measured directly. Existing works differ in what embedding Space is used and how to project the visual data into the semantic embedding Space. Yet, they all measure the similarity in the Space using a conventional distance metric (e.g. cosine) that does not consider the rich intrinsic structure, i.e. semantic manifold, of the semantic categories in the embedding Space. In this paper we propose to model the semantic manifold in an embedding Space using a semantic class label graph. The semantic manifold structure is used to redefine the distance metric in the semantic embedding Space for more effective ZSL. The proposed semantic manifold distance is computed using a novel absorbing Markov chain process (AMP), which has a very efficient closed-form solution. The proposed new model improves upon and seamlessly unifies various existing ZSL algorithms. Extensive experiments on both the large scale ImageNet dataset and the widely used Animal with Attribute (AwA) dataset show that our model outperforms significantly the state-of-the-arts.
-
cumulative Attribute Space for age and crowd density estimation
Computer Vision and Pattern Recognition, 2013Co-Authors: Ke Chen, Shaogang Gong, Tao XiangAbstract:A number of computer vision problems such as human age estimation, crowd density estimation and body/face pose (view angle) estimation can be formulated as a regression problem by learning a mapping function between a high dimensional vector-formed feature input and a scalar-valued output. Such a learning problem is made difficult due to sparse and imbalanced training data and large feature variations caused by both uncertain viewing conditions and intrinsic ambiguities between observable visual features and the scalar values to be estimated. Encouraged by the recent success in using Attributes for solving classification problems with sparse training data, this paper introduces a novel cumulative Attribute concept for learning a regression model when only sparse and imbalanced data are available. More precisely, low-level visual features extracted from sparse and imbalanced image samples are mapped onto a cumulative Attribute Space where each dimension has clearly defined semantic interpretation (a label) that captures how the scalar output value (e.g. age, people count) changes continuously and cumulatively. Extensive experiments show that our cumulative Attribute framework gains notable advantage on accuracy for both age estimation and crowd counting when compared against conventional regression models, especially when the labelled training data is sparse with imbalanced sampling.
