The Experts below are selected from a list of 3822 Experts worldwide ranked by ideXlab platform
Akira Taguchi - One of the best experts on this subject based on the ideXlab platform.
-
improvement of the image enlargement method based on the laplacian Pyramid Representation
IEEE-Eurasip Nonlinear Signal and Image Processing, 2005Co-Authors: A Watanabe, Akira TaguchiAbstract:Summary form only given. It is necessary to predict unknown higher-frequency components that are lost by sampling for enlarging digital images. Based on the Laplacian Pyramid Representation, the prediction of unknown higher-frequency components is equivalent to the prediction of a known higher-resolution Laplacian image. We have proposed the higher resolution method based on the Laplacian Pyramid Representation. However, the Laplacian Pyramid Representation was considered for image compression. Thus, we think that the bandwidth of the Gaussian filter for image compression is not optimal for digital image enlargement. In this paper, we proposed a new enlargement method for digital images with the variable bandwidth of Gaussian filter.
-
parameter estimations for super resolution based on the laplacian Pyramid Representation
European Signal Processing Conference, 2004Co-Authors: Shuai Yuan, Yasumasa Takahashi, Akira TaguchiAbstract:Image enlargement with higher frequency components prediction can be called as super resolution. In some print or copy applications, arbitrary scale enlargement is necessary. For obtaining better enlargement results, we have proposed one arbitrary scale super resolution method based on the Laplacian Pyramid Representation [5]. In this method there are two parameters in the higher frequency prediction processing. In fact, the exact relationship of them is very complex. To expound the meaning of these two parameters for arbitrary scale super resolution, in this paper we will do a mathematic analysis via the unit step signal. But we also point out that theoretical analysis maybe can't get the best effect for the natural image enlargement and in actual applications the experiential way should not be disregarded, too.
-
improvement of the image enlargement method based on the laplacian Pyramid Representation
Midwest Symposium on Circuits and Systems, 2004Co-Authors: A Watanabe, Akira TaguchiAbstract:The Laplacian Pyramid is the hierarchical expression. Based on Laplacian Pyramid Representation, the prediction of unknown higher-frequency components is equivalent to the prediction of an unknown high-resolution Laplacian image. Gaussian filter is used for calculating of the Laplacian Pyramid. And the band-width of the Gaussian filter is optimal for image compression. However, we cannot assert the band-width is optimal for the image enlargement method. In this paper, we change the band-width of the Gaussian filter in order to improve the performance of the enlargement method based on Laplacian Pyramid Representation. We show that the emphasis degree is changed by changing the band-width of the Gaussian filter.
-
an enlargement method of digital images based on laplacian Pyramid Representation
Electronics and Communications in Japan Part Ii-electronics, 2001Co-Authors: Yasumasa Takahashi, Akira TaguchiAbstract:Since the enlargement of digital images is carried out by insertion of sample points, the Nyquist frequency of the enlarged image is naturally larger than that prior to enlargement. Hence, estimation of unknown higher-frequency components is required for enlargement of digital images. On the other hand, Laplacian Pyramid Representation is widely known as a layered expression of the images. Based on this Laplacian Pyramid Representation, the unknown higher-frequency components to be estimated at the time of enlargement of digital images correspond to the unknown higher-resolution Laplacian components. There is strong correlation between the Laplacian images at different layers. Specifically, the edge information spans several continuous layers as zero crossings and hence appears at the same location. Differences in the layers are recognized as differences in gradient near the zero crossing. In this paper, this property is directly used to provide a method of obtaining unknown higher-resolution Laplacian images from low-resolution Laplacian images (direct method). The results of enlargement by the direct method are particularly good in the smoothly varying parts of the images, but there are some difficulties in enlargement at the edges and of details. The paper proposes a hybrid method in which the direct method is combined with the method of Greenspan, with excellent enlargement results at edges and details, after the enlargement of images by estimation of the unknown higher-resolution Laplacian images as in the direct method. It is demonstrated by many application examples that this hybrid method is reasonably simple and is superior numerically and subjectively. © 2001 Scripta Technica, Electron Comm Jpn Pt 2, 84(6): 40–49, 2001
-
enlargement method of digital images based on laplacian Pyramid Representation
electronic imaging, 2000Co-Authors: Akira Taguchi, Yasumasa TakahashiAbstract:We present novel enlarging methods of digital images based on Laplacian Pyramid Representation. First, 'direct method,' which is based on Laplacian Pyramid Representation, is proposed. The direct method is simple, however derive excellent enlargement results for only relative smooth regions of the image. Thus, we present the 'hybrid method' which is integrated the direct method and the Greenspan's method which is also based on Laplacian Pyramid Representation and has excellent property for enraging edge/detail regions. The effectiveness of the hybrid method is shown through a lot of experimental results.© (2000) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.
Yasumasa Takahashi - One of the best experts on this subject based on the ideXlab platform.
-
parameter estimations for super resolution based on the laplacian Pyramid Representation
European Signal Processing Conference, 2004Co-Authors: Shuai Yuan, Yasumasa Takahashi, Akira TaguchiAbstract:Image enlargement with higher frequency components prediction can be called as super resolution. In some print or copy applications, arbitrary scale enlargement is necessary. For obtaining better enlargement results, we have proposed one arbitrary scale super resolution method based on the Laplacian Pyramid Representation [5]. In this method there are two parameters in the higher frequency prediction processing. In fact, the exact relationship of them is very complex. To expound the meaning of these two parameters for arbitrary scale super resolution, in this paper we will do a mathematic analysis via the unit step signal. But we also point out that theoretical analysis maybe can't get the best effect for the natural image enlargement and in actual applications the experiential way should not be disregarded, too.
-
an enlargement method of digital images based on laplacian Pyramid Representation
Electronics and Communications in Japan Part Ii-electronics, 2001Co-Authors: Yasumasa Takahashi, Akira TaguchiAbstract:Since the enlargement of digital images is carried out by insertion of sample points, the Nyquist frequency of the enlarged image is naturally larger than that prior to enlargement. Hence, estimation of unknown higher-frequency components is required for enlargement of digital images. On the other hand, Laplacian Pyramid Representation is widely known as a layered expression of the images. Based on this Laplacian Pyramid Representation, the unknown higher-frequency components to be estimated at the time of enlargement of digital images correspond to the unknown higher-resolution Laplacian components. There is strong correlation between the Laplacian images at different layers. Specifically, the edge information spans several continuous layers as zero crossings and hence appears at the same location. Differences in the layers are recognized as differences in gradient near the zero crossing. In this paper, this property is directly used to provide a method of obtaining unknown higher-resolution Laplacian images from low-resolution Laplacian images (direct method). The results of enlargement by the direct method are particularly good in the smoothly varying parts of the images, but there are some difficulties in enlargement at the edges and of details. The paper proposes a hybrid method in which the direct method is combined with the method of Greenspan, with excellent enlargement results at edges and details, after the enlargement of images by estimation of the unknown higher-resolution Laplacian images as in the direct method. It is demonstrated by many application examples that this hybrid method is reasonably simple and is superior numerically and subjectively. © 2001 Scripta Technica, Electron Comm Jpn Pt 2, 84(6): 40–49, 2001
-
enlargement method of digital images based on laplacian Pyramid Representation
electronic imaging, 2000Co-Authors: Akira Taguchi, Yasumasa TakahashiAbstract:We present novel enlarging methods of digital images based on Laplacian Pyramid Representation. First, 'direct method,' which is based on Laplacian Pyramid Representation, is proposed. The direct method is simple, however derive excellent enlargement results for only relative smooth regions of the image. Thus, we present the 'hybrid method' which is integrated the direct method and the Greenspan's method which is also based on Laplacian Pyramid Representation and has excellent property for enraging edge/detail regions. The effectiveness of the hybrid method is shown through a lot of experimental results.© (2000) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.
Wei Feng - One of the best experts on this subject based on the ideXlab platform.
-
the spatial laplacian and temporal energy Pyramid Representation for human action recognition using depth sequences
Knowledge Based Systems, 2017Co-Authors: Jun Cheng, Dapeng Tao, Wei FengAbstract:Depth sequences are useful for action recognition since they are insensitive to illumination variation and provide geometric information. Many current action recognition methods are limited by being computationally expensive and requiring large-scale training data. Here we propose an effective method for human action recognition using depth sequences captured by depth cameras. A multi-resolution operation, the spatial Laplacian and temporal energy Pyramid (SLTEP), decomposes the depth sequences into certain frequency bands in different space and time positions. A spatial aggregating and fusion scheme is applied to cluster the low-level features and concatenate two different feature types extracted from low and high frequency levels, respectively. We evaluate our approach on five public benchmark datasets (MSRAction3D, MSRGesture3D, MSRActionPairs, MSRDailyActivity3D, and NTU RGB+D) and demonstrate its advantages over existing methods and is likely to be highly useful for online applications.
Stephen Lin - One of the best experts on this subject based on the ideXlab platform.
-
beyond spatial Pyramids a new feature extraction framework with dense spatial sampling for image classification
European Conference on Computer Vision, 2012Co-Authors: Shengye Yan, Stephen LinAbstract:We introduce a new framework for image classification that extends beyond the window sampling of fixed spatial Pyramids to include a comprehensive set of windows densely sampled over location, size and aspect ratio. To effectively deal with this large set of windows, we derive a concise high-level image feature using a two-level extraction method. At the first level, window-based features are computed from local descriptors (e.g., SIFT, spatial HOG, LBP) in a process similar to standard feature extractors. Then at the second level, the new image feature is determined from the window-based features in a manner analogous to the first level. This higher level of abstraction offers both efficient handling of dense samples and reduced sensitivity to misalignment. More importantly, our simple yet effective framework can readily accommodate a large number of existing pooling/coding methods, allowing them to extract features beyond the spatial Pyramid Representation. To effectively fuse the second level feature with a standard first level image feature for classification, we additionally propose a new learning algorithm, called Generalized Adaptive lp-norm Multiple Kernel Learning (GA-MKL), to learn an adapted robust classifier based on multiple base kernels constructed from image features and multiple sets of pre-learned classifiers of all the classes. Extensive evaluation on the object recognition (Caltech256) and scene recognition (15Scenes) benchmark datasets demonstrates that the proposed method outperforms state-of-the-art image classification algorithms under a broad range of settings.
Minghsuan Yang - One of the best experts on this subject based on the ideXlab platform.
-
hierarchical convolutional features for visual tracking
International Conference on Computer Vision, 2015Co-Authors: Jiabin Huang, Xiaokang Yang, Minghsuan YangAbstract:Visual object tracking is challenging as target objects often undergo significant appearance changes caused by deformation, abrupt motion, background clutter and occlusion. In this paper, we exploit features extracted from deep convolutional neural networks trained on object recognition datasets to improve tracking accuracy and robustness. The outputs of the last convolutional layers encode the semantic information of targets and such Representations are robust to significant appearance variations. However, their spatial resolution is too coarse to precisely localize targets. In contrast, earlier convolutional layers provide more precise localization but are less invariant to appearance changes. We interpret the hierarchies of convolutional layers as a nonlinear counterpart of an image Pyramid Representation and exploit these multiple levels of abstraction for visual tracking. Specifically, we adaptively learn correlation filters on each convolutional layer to encode the target appearance. We hierarchically infer the maximum response of each layer to locate targets. Extensive experimental results on a largescale benchmark dataset show that the proposed algorithm performs favorably against state-of-the-art methods.
