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

Computational single-cell RNA-seq (scRNA-seq) methods have been successfully applied to experiments representing a single condition, technology, or species to discover and define cellular phenotypes. However, identifying subpopulations of cells that are present across multiple data sets remains challenging. Here, we introduce an analytical strategy for integrating scRNA-seq data sets based on common sources of variation, enabling the identification of shared populations across data sets and downstream comparative analysis. We apply this approach, implemented in our R toolkit Seurat (http://satijalab.org/seurat/), to align scRNA-seq data sets of peripheral blood mononuclear cells under resting and stimulated conditions, hematopoietic progenitors sequenced using two profiling technologies, and pancreatic cell 'atlases' generated from human and mouse islets. In each case, we learn distinct or transitional cell states jointly across data sets, while boosting statistical power through integrated analysis. Our approach facilitates general comparisons of scRNA-seq data sets, potentially deepening our understanding of how distinct cell states respond to perturbation, disease, and evolution.

/pdf/integrating-single-cell-transcriptomic-data-across-different-3dtsi5w7ob.pdf

Integrating single-cell transcriptomic data across different conditions, technologies, and species.

/pdf/convex-analysisnoer-san-nojin-zhan-nituite-5dl2rbmoyr.pdf

Convex Analysisの二,三の進展について

This paper presents a novel and efficient facial image representation based on local binary pattern (LBP) texture features. The face image is divided into several regions from which the LBP feature distributions are extracted and concatenated into an enhanced feature vector to be used as a face descriptor. The performance of the proposed method is assessed in the face recognition problem under different challenges. Other applications and several extensions are also discussed

Face Description with Local Binary Patterns: Application to Face Recognition

“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

We propose a new task named Audio-driven Per-formance Video Generation (APVG), which aims to synthesizethe video of a person playing a certain instrument guided bya given music audio clip. It is a challenging task to gener-ate the high-dimensional temporal consistent videos from low-dimensional audio modality. In this paper, we propose a multi-staged framework to achieve this new task to generate realisticand synchronized performance video from given music. Firstly,we provide both global appearance and local spatial informationby generating the coarse videos and keypoints of body and handsfrom a given music respectively. Then, we propose to transformthe generated keypoints to heatmap via a differentiable spacetransformer, since the heatmap offers more spatial informationbut is harder to generate directly from audio. Finally, wepropose a Structured Temporal UNet (STU) to extract bothintra-frame structured information and inter-frame temporalconsistency. They are obtained via graph-based structure module,and CNN-GRU based high-level temporal module respectively forfinal video generation. Comprehensive experiments validate theeffectiveness of our proposed framework.

Lets Play Music: Audio-driven Performance Video Generation

This paper introduces a novel convex kernel based method for color constancy computation with explicit illuminant parameter estimation. A simple linear render model is adopted and the illuminants in a new scene that contains some of the color surfaces seen in the training image are sequentially estimated in a global optimization framework. The proposed method is fully data-driven and initialization invariant. Nonlinear color constancy can also be approximately solved in this kernel optimization framework with piecewise linear assumption. Extensive experiments on real-scene images validate the practical performance of our method.

/pdf/color-constancy-via-convex-kernel-optimization-4yjy1ea5i6.pdf

Color constancy via convex kernel optimization

Automatic, fast and accurate extraction of the blood vessel is an important task in the image-aided diagnosis of disease. In this paper, we describe a novel superficial vessel imaging and projecting system. First, the superficial vessel of human arm is captured by NIR imaging technique, and then three pre-processing algorithms, including hair removal, non-uniform illumination correction and vessel enhancement, are developed to strengthen the vessel image. Second, a model-based binarization method is proposed to detect vessel-like structures. And then the mathematic morphological and connected component refinement methods are integrated to remove small vessel noises. Experimental results demonstrate that our system can imaging superficial vessels on real-time. The proposed system does not need any human interaction, which hence can be used in clinical venipuncture practices.

Advances in Image and Graphics Technologies

This chapter gives an overview of the heterogeneous problem in facial analysis as well as its synthesis solution, followed by a brief outline of the rest chapters. We start from the background and challenges in research of heterogeneous facial analysis. Then the heterogeneous facial synthesis is emphasized as one of the promising and effective solutions. The instantiated tasks to be elaborated in the following chapters are finally introduced in a compact manner.

Heterogeneous Facial Analysis and Synthesis

In this paper, we introduce an information theory motivated algorithm for constructing a low dimensional representation for data sampled from a higher dimensional space. The proposed minimum entropy linear embedding algorithm tries to minimize the information uncertainty (measured by entropy) as much as possible. The entropy is estimated by Gaussian mixture model probability density function and an upper bound of entropy is derived. As a result, the numerical integration involved in the objective function is reduced to a computationally efficient eigenfunction problem. The superiority of proposed method is that it can be used to find the intrinsic character of high dimensional data and has potential ability to reduce redundancy and to improve classification accuracy. Numerical results on toy data, UCI machine learning data set and face recognition illustrate this superiority.

Ran He

Papers

Lets Play Music: Audio-driven Performance Video Generation

Color constancy via convex kernel optimization

Advances in Image and Graphics Technologies

Heterogeneous Facial Analysis and Synthesis

Minimum entropy linear embedding based on Gaussian mixture model