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

Near-infrared-visible (NIR-VIS) heterogeneous face recognition matches NIR to corresponding VIS face images. However, due to the sensing gap, NIR images often lose some identity information so that the recognition issue is more difficult than conventional VIS face recognition. Recently, NIR-VIS heterogeneous face recognition has attracted considerable attention in the computer vision community because of its convenience and adaptability in practical applications. Various deep learning-based methods have been proposed and substantially increased the recognition performance, but the lack of NIR-VIS training samples leads to the difficulty of the model training process. In this paper, we propose a new Large-Scale Multi-Pose High-Quality NIR-VIS database LAMP-HQ containing 56,788 NIR and 16,828 VIS images of 573 subjects with large diversities in pose, illumination, attribute, scene and accessory. We furnish a benchmark along with the protocol for NIR-VIS face recognition via generation on LAMP-HQ, including Pixel2Pixel, CycleGAN, and ADFL. Furthermore, we propose a novel exemplar-based variational spectral attention network to produce high-fidelity VIS images from NIR data. A spectral conditional attention module is introduced to reduce the domain gap between NIR and VIS data and then improve the performance of NIR-VIS heterogeneous face recognition on various databases including the LAMP-HQ.

LAMP-HQ: A Large-Scale Multi-Pose High-Quality Database and Benchmark for NIR-VIS Face Recognition

Sparse signal representation proposes a novel insight to solve face recognition problem. Based on the sparse assumption that a new object can be sparsely represented by other objects, we propose a simple yet efficient direct sparse nearest feature classifier to deal with the problem of automatically real-time face recognition. Firstly, we present a new method, which calculates an approximate sparse code to alleviate the extrapolation and interpolation inaccuracy in nearest feature classifier. Secondly, a sparse score normalization method is developed to normalize the calculated scores and to achieve a high receiver operator characteristic (ROC) curve. Experiments on FRGC and PIE face databases show that our method can get comparable results against sparse representation-based classification on both recognition rate and ROC curve.

Direct sparse nearest feature classifier for face recognition

Digital recognition with lip images has become a key step of the interactive liveness detection for Chinese banking systems. However, the problem of the digital recognition is very challenging due to intra class variation of lip images, head pose variations, and uncontrolled illumination. This paper studies a deep learning architecture to model the appearance and the spatial-temporal information of lip texture. The lip texture in still image frames and the spatial-temporal relationship between these frames are jointly modeled by convolutional neural networks and long short-term memory. Two strategies are further exploited to find effective groups of ten digitals for training the deep models. As a result, more information can be utilized for accurate recognition based on lip texture analysis. Besides, two datasets of isolated digits in Chinese are established to simulate real-world liveness detection environments together with various attacks. Extensive experiments have been done to analyze the recognition accuracy of each digit and to provide some clues for determining appropriate digits for interactive liveness detection.

Digital recognition from lip texture analysis

Combinatorial optimization (CO) has been a hot research topic because of its theoretic and practical importance. As a classic CO problem, deep hashing aims to find an optimal code for each data from finite discrete possibilities, while the discrete nature brings a big challenge to the optimization process. Previous methods usually mitigate this challenge by binary approximation, substituting binary codes for real-values via activation functions or regularizations. However, such approximation leads to uncertainty between real-values and binary ones, degrading retrieval performance. In this paper, we propose a novel Deep Momentum Uncertainty Hashing (DMUH). It explicitly estimates the uncertainty during training and leverages the uncertainty information to guide the approximation process. Specifically, we model bit-level uncertainty via measuring the discrepancy between the output of a hashing network and that of a momentum-updated network. The discrepancy of each bit indicates the uncertainty of the hashing network to the approximate output of that bit. Meanwhile, the mean discrepancy of all bits in a hashing code can be regarded as image-level uncertainty. It embodies the uncertainty of the hashing network to the corresponding input image. The hashing bit and image with higher uncertainty are paid more attention during optimization. To the best of our knowledge, this is the first work to study the uncertainty in hashing bits. Extensive experiments are conducted on four datasets to verify the superiority of our method, including CIFAR-10, NUS-WIDE, MS-COCO, and a million-scale dataset Clothing1M. Our method achieves the best performance on all of the datasets and surpasses existing state-of-the-art methods by a large margin.

Deep Momentum Uncertainty Hashing

To narrow the inherent gap between heterogeneous face domains, much efforts have been made in the past decade Some of these solutions resort to deep generative networks to synthesize proper face images based on inputs from another domain It is widely agreed that facial topology tends to be closely related to identity information, since facial appearance has been deemed as a crucial feature for identifying individuals A small topological change on the face may make one person become another Taking makeup removal as an example, the synthesized output is expected to resemble the input except for the makeup effects, and the facial topology is premier to be preserved during synthesis We generalize this kind of problem as topology-invariant facial synthesis In this chapter, we select several representative topology-invariant synthesis tasks and elaborate their latest progress These tasks include cross-spectral face hallucination, makeup removal and age progression/regression For each task, we start from its background and characteristics, followed by the illustration of latest methods as well as their experimental outcomes

Ran He

Papers

LAMP-HQ: A Large-Scale Multi-Pose High-Quality Database and Benchmark for NIR-VIS Face Recognition

Direct sparse nearest feature classifier for face recognition

Digital recognition from lip texture analysis

Deep Momentum Uncertainty Hashing

Topology-Invariant Synthesis