The Experts below are selected from a list of 132693 Experts worldwide ranked by ideXlab platform
Tao Yang - One of the best experts on this subject based on the ideXlab platform.
-
ConvNet and LSH-Based Visual Localization Using Localized Sequence Matching
Sensors, 2019Co-Authors: Yongliang Qiao, Cindy Cappelle, Yassine Ruichek, Tao YangAbstract:Convolutional Network (ConvNet), with its strong Image representation ability, has achieved significant progress in the computer vision and robotic fields. In this paper, we propose a visual localization approach based on place recognition that combines the powerful ConvNet features and localized Image sequence matching. The Image Distance matrix is constructed based on the cosine Distance of extracted ConvNet features, and then a sequence search technique is applied on this Distance matrix for the final visual recognition. To speed up the computational efficiency, the locality sensitive hashing (LSH) method is applied to achieve real-time performances with minimal accuracy degradation. We present extensive experiments on four real world data sets to evaluate each of the specific challenges in visual recognition. A comprehensive performance comparison of different ConvNet layers (each defining a level of features) considering both appearance and illumination changes is conducted. Compared with the traditional approaches based on hand-crafted features and single Image matching, the proposed method shows good performances even in the presence of appearance and illumination changes.
Yongliang Qiao - One of the best experts on this subject based on the ideXlab platform.
-
ConvNet and LSH-Based Visual Localization Using Localized Sequence Matching
Sensors, 2019Co-Authors: Yongliang Qiao, Cindy Cappelle, Yassine Ruichek, Tao YangAbstract:Convolutional Network (ConvNet), with its strong Image representation ability, has achieved significant progress in the computer vision and robotic fields. In this paper, we propose a visual localization approach based on place recognition that combines the powerful ConvNet features and localized Image sequence matching. The Image Distance matrix is constructed based on the cosine Distance of extracted ConvNet features, and then a sequence search technique is applied on this Distance matrix for the final visual recognition. To speed up the computational efficiency, the locality sensitive hashing (LSH) method is applied to achieve real-time performances with minimal accuracy degradation. We present extensive experiments on four real world data sets to evaluate each of the specific challenges in visual recognition. A comprehensive performance comparison of different ConvNet layers (each defining a level of features) considering both appearance and illumination changes is conducted. Compared with the traditional approaches based on hand-crafted features and single Image matching, the proposed method shows good performances even in the presence of appearance and illumination changes.
J. Rousseau - One of the best experts on this subject based on the ideXlab platform.
-
assessment of Image intensifier and distortion for dsa localization studies
British Journal of Radiology, 1997Co-Authors: EMMANUEL COSTE, D Gibon, J. RousseauAbstract:The authors present methods of correcting pincushion and S distortions of an Image intensifier, and of measuring the geometrical parameters of the imaging device used for localization from digital subtraction angiography brain studies. Pincushion and S distortions of the Image intensifier are corrected by a calibration grid. A test pattern is used to study effectiveness of the corrections. Intrinsic geometrical parameters (source-to-Image Distance, centre of X-ray projection) of the apparatus are measured by the use of a calibration phantom. Short-range and long-range time drift of the distortion, as well as influence of the parameters of Image acquisition on the accuracy of the localization results, are considered. The results obtained successfully demonstrate the accuracy of the correction, provided that the apparatus is warm.
Chunhua Shen - One of the best experts on this subject based on the ideXlab platform.
-
deepemd few shot Image classification with differentiable earth mover s Distance and structured classifiers
Computer Vision and Pattern Recognition, 2020Co-Authors: Chi Zhang, Yujun Cai, Guosheng Lin, Chunhua ShenAbstract:In this paper, we address the few-shot classification task from a new perspective of optimal matching between Image regions. We adopt the Earth Mover's Distance (EMD) as a metric to compute a structural Distance between dense Image representations to determine Image relevance. The EMD generates the optimal matching flows between structural elements that have the minimum matching cost, which is used to represent the Image Distance for classification. To generate the important weights of elements in the EMD formulation, we design a cross-reference mechanism, which can effectively minimize the impact caused by the cluttered background and large intra-class appearance variations. To handle k-shot classification, we propose to learn a structured fully connected layer that can directly classify dense Image representations with the EMD. Based on the implicit function theorem, the EMD can be inserted as a layer into the network for end-to-end training. We conduct comprehensive experiments to validate our algorithm and we set new state-of-the-art performance on four popular few-shot classification benchmarks, namely miniImageNet, tieredImageNet, Fewshot-CIFAR100 (FC100) and Caltech-UCSD Birds-200-2011 (CUB).
-
deepemd few shot Image classification with differentiable earth mover s Distance and structured classifiers
2020Co-Authors: Chi Zhang, Yujun Cai, Guosheng Lin, Chunhua ShenAbstract:Deep learning has proved to be very effective in learning with a large amount of labelled data. Few-shot learning in contrast attempts to learn with only a few labelled data. In this work, we develop methods for few-shot Image classification from a new perspective of optimal matching between Image regions. We employ the Earth Mover's Distance (EMD) as a metric to compute a structural Distance between dense Image representations to determine Image relevance. The EMD generates the optimal matching flows between structural elements that have the minimum matching cost, which is used to calculate the Image Distance for classification. To generate the important weights of elements in the EMD formulation, we design a cross-reference mechanism, which can effectively alleviate the adverse impact caused by the cluttered background and large intra-class appearance variations. To handle $k$-shot classification, we propose to learn a structured fully connected layer that can directly classify dense Image representations with the proposed EMD. Based on the implicit function theorem, the EMD can be inserted as a layer into the network for end-to-end training. Our extensive experiments validate the effectiveness of our algorithm which outperforms state-of-the-art methods by a significant margin on four widely used few-shot classification benchmarks, namely, miniImageNet, tieredImageNet, Fewshot-CIFAR100 (FC100) and Caltech-UCSD Birds-200-2011 (CUB).
-
deepemd differentiable earth mover s Distance for few shot learning
arXiv: Computer Vision and Pattern Recognition, 2020Co-Authors: Chi Zhang, Yujun Cai, Guosheng Lin, Chunhua ShenAbstract:Deep learning has proved to be very effective in learning with a large amount of labelled data. Few-shot learning in contrast attempts to learn with only a few labelled data. In this work, we develop methods for few-shot Image classification from a new perspective of optimal matching between Image regions. We employ the Earth Mover's Distance (EMD) as a metric to compute a structural Distance between dense Image representations to determine Image relevance. The EMD generates the optimal matching flows between structural elements that have the minimum matching cost, which is used to calculate the Image Distance for classification. To generate the important weights of elements in the EMD formulation, we design a cross-reference mechanism, which can effectively alleviate the adverse impact caused by the cluttered background and large intra-class appearance variations. To handle $k$-shot classification, we propose to learn a structured fully connected layer that can directly classify dense Image representations with the proposed EMD. Based on the implicit function theorem, the EMD can be inserted as a layer into the network for end-to-end training. Our extensive experiments validate the effectiveness of our algorithm which outperforms state-of-the-art methods by a significant margin on four widely used few-shot classification benchmarks, namely, miniImageNet, tieredImageNet, Fewshot-CIFAR100 (FC100) and Caltech-UCSD Birds-200-2011 (CUB).
Cindy Cappelle - One of the best experts on this subject based on the ideXlab platform.
-
ConvNet and LSH-Based Visual Localization Using Localized Sequence Matching
Sensors, 2019Co-Authors: Yongliang Qiao, Cindy Cappelle, Yassine Ruichek, Tao YangAbstract:Convolutional Network (ConvNet), with its strong Image representation ability, has achieved significant progress in the computer vision and robotic fields. In this paper, we propose a visual localization approach based on place recognition that combines the powerful ConvNet features and localized Image sequence matching. The Image Distance matrix is constructed based on the cosine Distance of extracted ConvNet features, and then a sequence search technique is applied on this Distance matrix for the final visual recognition. To speed up the computational efficiency, the locality sensitive hashing (LSH) method is applied to achieve real-time performances with minimal accuracy degradation. We present extensive experiments on four real world data sets to evaluate each of the specific challenges in visual recognition. A comprehensive performance comparison of different ConvNet layers (each defining a level of features) considering both appearance and illumination changes is conducted. Compared with the traditional approaches based on hand-crafted features and single Image matching, the proposed method shows good performances even in the presence of appearance and illumination changes.
