[신간의 별자리x] 우리/미술, 그리고 ‘슬픔의 박물관’

Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

I have developed "tennis elbow" from lugging this book around the past four weeks, but it is worth the pain, the effort, and the aspirin. It is also worth the (relatively speaking) bargain price. Including appendixes, this book contains 894 pages of text. The entire panorama of the neural sciences is surveyed and examined, and it is comprehensive in its scope, from genomes to social behaviors. The editors explicitly state that the book is designed as "an introductory text for students of biology, behavior, and medicine," but it is hard to imagine any audience, interested in any fragment of neuroscience at any level of sophistication, that would not enjoy this book. The editors have done a masterful job of weaving together the biologic, the behavioral, and the clinical sciences into a single tapestry in which everyone from the molecular biologist to the practicing psychiatrist can find and appreciate his or

Principles of Neural Science

Principal component analysis is one of the most important and powerful methods in chemometrics as well as in a wealth of other areas. This paper provides a description of how to understand, use, and interpret principal component analysis. The paper focuses on the use of principal component analysis in typical chemometric areas but the results are generally applicable.

/pdf/principal-component-analysis-1sba76ou47.pdf

Principal component analysis

Photoplethysmography and its application in clinical physiological measurement

Deep convolutional networks provide very high quality super resolution images through a learning process by a nonlinear end-to-end mapping between low and high resolution images. Many of the state-of-the-art super resolution networks employ residual blocks in their network architectures, where in each residual block the high frequency residual signals are added to the feature maps input to the block. In this paper, a new residual block is proposed for the problem of image super resolution. The proposed residual block consists of three modules, namely, feature transformation module, nonlinear edge extraction module and feature fusion module. The feature transformation module produces high frequency residual signals and the nonlinear edge extraction module extracts the edges of the features input to the block. These generated high frequency features are then fused using the feature fusion module in order to produce a very rich set of high frequency residual features. The performance of the super resolution network using the proposed residual block is compared with that of the state-of-the-art light-weight super resolution schemes on four benchmark datasets. It is shown that the proposed super resolution scheme outperforms the state-of-the-art light-weight super resolution networks, when both the performance and number of parameters of the network are simultaneously taken into consideration.

EFFRBNet: A Deep Super Resolution Network using Edge-Assisted Feature Fusion Residual Blocks

In this paper, a structural local DCT sparse appearance model with occlusion detection is proposed for visual tracking in a particle filter framework. The energy compaction property of the 2D-DCT is exploited to reduce the size of the dictionary as well as that of the candidate samples so that the computational cost of l1-minimization can be lowered. Further, a holistic image reconstruction procedure is proposed for robust occlusion detection and used for appearance model update, thus avoiding the degradation of the appearance model in the presence of occlusion/outliers. Also, a patch occlusion ratio is introduced in the confidence score computation to enhance the tracking performance. Quantitative and qualitative performance evaluations on two popular benchmark datasets demonstrate that the proposed tracking algorithm generally outperforms several state-of-the-art methods.

/pdf/visual-tracking-using-structural-local-dct-sparse-appearance-2ix0fafwuj.pdf

Visual tracking using structural local DCT sparse appearance model with occlusion detection

In feature-level fusion, features extracted from different modalities are fused in order to obtain a single feature set for multimodal biometric recognition systems. These features can be encoded using a binary (1' or '0') encoding technique. The encoded feature value of '1' provides more information about the feature than '0' does. In view of this, we first propose a fusion in order to fuse encoded features obtained from individual feature encoders for a multimodal biometric system, and refer to it as the first-stage fusion (FSF). Next, another fusion is carried out between the unimodal system which provides the best performance in that multimodal system and the proposed FSF, and referred to as the second-stage fusion (SSF). Genuine acceptance rates @4.3% and @4.4% false acceptance rates, and equal error rate are utilized for evaluating the performance of a multi-biometric system using the proposed fusions. Results show that a superior performance is provided by a multi-biometric system using the proposed fusion scheme in comparison with the performance provided by the system using existing fusions or by the unimodal systems.

A Two-Stage Scheme for Fusion of Hash-Encoded Features in a Multimodal Biometric System

In this paper, a pyramidal motion estimation technique that makes use of the motion correlation within a pyramidal level is proposed. Instead of scaling the motion vectors from the adjacent lower pyramidal level as motion predictions as is done in the conventional technique, in the proposed technique, the motion vectors from the neighboring motion blocks are taken into consideration as possible candidates. Each of these candidate motion vectors is used for prediction and refined. The motion vector that has the least matching distortion is chosen as the final motion vector. Compared to the conventional pyramidal motion estimation technique, the proposed method effectively overcomes the problem of propagation of false motion vectors. Simulation studies show that a substantial performance improvement is achieved, both in terms of the prediction mean square error and the number of coding bits for the motion vectors.

A neighborhood-blocks motion estimation technique using the pyramidal data structure

Use of large image datasets has become a common occurrence. This, however, makes image searching a highly desired operation in many applications. Most of the content-based image retrieval (CBIR) methods usually adopt machine-learning techniques that take the image content into account. These methods are effective, but they are generally too complex and resource demanding. We propose a framework based on image hashing and random forest, which is fast and offers high performance. The proposed framework consists of a multi-key image hashing technique based on discrete cosine transform (DCT) and discrete wavelet transform (DWT) and random forest based on normalized B+ Tree (NB+ Tree), which reduces the high-dimensional input vectors to one-dimension, which in turn improves the time complexity significantly. We analyze our method empirically and show that it outperforms competitive methods in terms of both accuracy and speed. In addition, the proposed scheme maintains a fast scaling with increasing size of the data sets while preserving high accuracy.

M. Omair Ahmad

Papers

EFFRBNet: A Deep Super Resolution Network using Edge-Assisted Feature Fusion Residual Blocks

Visual tracking using structural local DCT sparse appearance model with occlusion detection

A Two-Stage Scheme for Fusion of Hash-Encoded Features in a Multimodal Biometric System

A neighborhood-blocks motion estimation technique using the pyramidal data structure

Perceptual Image Hashing Using Random Forest for Content-based Image Retrieval