We present YOLO, a new approach to object detection. Prior work on object detection repurposes classifiers to perform detection. Instead, we frame object detection as a regression problem to spatially separated bounding boxes and associated class probabilities. A single neural network predicts bounding boxes and class probabilities directly from full images in one evaluation. Since the whole detection pipeline is a single network, it can be optimized end-to-end directly on detection performance. Our unified architecture is extremely fast. Our base YOLO model processes images in real-time at 45 frames per second. A smaller version of the network, Fast YOLO, processes an astounding 155 frames per second while still achieving double the mAP of other real-time detectors. Compared to state-of-the-art detection systems, YOLO makes more localization errors but is less likely to predict false positives on background. Finally, YOLO learns very general representations of objects. It outperforms other detection methods, including DPM and R-CNN, when generalizing from natural images to other domains like artwork.

/pdf/you-only-look-once-unified-real-time-object-detection-3gan6roe1o.pdf

You Only Look Once: Unified, Real-Time Object Detection

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

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

We consider the problem of automatically recognizing human faces from frontal views with varying expression and illumination, as well as occlusion and disguise. We cast the recognition problem as one of classifying among multiple linear regression models and argue that new theory from sparse signal representation offers the key to addressing this problem. Based on a sparse representation computed by C1-minimization, we propose a general classification algorithm for (image-based) object recognition. This new framework provides new insights into two crucial issues in face recognition: feature extraction and robustness to occlusion. For feature extraction, we show that if sparsity in the recognition problem is properly harnessed, the choice of features is no longer critical. What is critical, however, is whether the number of features is sufficiently large and whether the sparse representation is correctly computed. Unconventional features such as downsampled images and random projections perform just as well as conventional features such as eigenfaces and Laplacianfaces, as long as the dimension of the feature space surpasses certain threshold, predicted by the theory of sparse representation. This framework can handle errors due to occlusion and corruption uniformly by exploiting the fact that these errors are often sparse with respect to the standard (pixel) basis. The theory of sparse representation helps predict how much occlusion the recognition algorithm can handle and how to choose the training images to maximize robustness to occlusion. We conduct extensive experiments on publicly available databases to verify the efficacy of the proposed algorithm and corroborate the above claims.

/pdf/robust-face-recognition-via-sparse-representation-4bfqv6y74o.pdf

Robust Face Recognition via Sparse Representation

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.

Manifold learning (ML) aims to seek low-dimensional embedding from high-dimensional data. The problem is challenging on real-world datasets, especially with under-sampling data, and we find that previous methods perform poorly in this case. Generally, ML methods first transform input data into a low-dimensional embedding space to maintain the data’s geometric structure and subsequently perform downstream tasks therein. The poor local connectivity of under-sampling data in the former step and inappropriate optimization objectives in the latter step leads to two problems: structural distortion and underconstrained embedding. This paper proposes a novel ML framework named Deep Local-flatness Manifold Embedding (DLME) to solve these problems. The proposed DLME constructs semantic manifolds by data augmentation and overcomes the structural distortion problem using a smoothness constrained based on a local flatness assumption about the manifold. To overcome the underconstrained embedding problem, we design a loss and theoretically demonstrate that it leads to a more suitable embedding based on the local flatness. Experiments on three types of datasets (toy, biological, and image) for various downstream tasks (classification, clustering, and visualization) show that our proposed DLME outperforms state-of-the-art ML and contrastive learning methods. 

DLME: Deep Local-Flatness Manifold Embedding

Although face detection has taken a big step forward with the development of anchor based face detector, the issue of effective detection of faces with different scales still remains. To solve this problem, we present an one-stage face detector, named Single Shot Attention-Based Face Detector (AFD), which enables accurate detection of multi-scale faces with high efficiency, especially for small faces. Specifically, AFD consists of two inter-connected modules, namely attention proposal module (APM) and face detection module (FDM). The former aims to generate the attention region and coarsely refine the anchors. The latter takes the output from APM as input and further improve the detection results. We obtain state-of-the-art results on common face detection benchmarks, i.e. FDDB and WIDER FACE, and can run at 20 FPS on a Nvidia Titan X (Pascal) for VGA-resolution images.

Single Shot Attention-Based Face Detector

A main problem in face retrieval is the semantic gap between low-level features and high-level semantic concepts. Relevance feedback (RF) may be used to incorporate to reduce the semantic gap. However, in the search for a specific target in a facial image database, a user's assignment of RF instances may be mistaken. This would make the system prediction of the user's target in a wrong way. Addressing this problem, we propose a new query point movement technique for target search by posing the problem of reducing the impact of inaccurate user feedback as an optimization problem. We develop a support vector machine based method to learn a decision boundary to identify ideal irrelevant images. Then we propose a rank function for finding target images, which would assign high scores to the images near the relevant images and punish those close to the decision boundary. Experiments are performed to show the stability and efficiency of the proposed algorithm.

Reducing Impact of Inaccurate User Feedback in Face Retrieval

Regardless of the usage of deep learning and handcrafted methods, the dynamic information from videos and the effect of cross-ethnicity are rarely considered in face anti-spoofing. In this work, we propose a static-dynamic fusion mechanism for multi-modal face anti-spoofing. Inspired by motion divergences between real and fake faces, we incorporate the dynamic image calculated by rank pooling with static information into a conventional neural network (CNN) for each modality (i.e., RGB, Depth and infrared (IR)). Then, we develop a partially shared fusion method to learn complementary information from multiple modalities. Furthermore, in order to study the generalization capability of the proposal in terms of cross-ethnicity attacks and unknown spoofs, we introduce the largest public cross-ethnicity Face Anti-spoofing (CASIA-CeFA) dataset, covering 3 ethnicities, 3 modalities, 1607 subjects, and 2D plus 3D attack types. Experiments demonstrate that the proposed method achieves state-of-the-art results on CASIA-CeFA, CASIA-SURF, OULU-NPU and SiW.

Static and Dynamic Fusion for Multi-modal Cross-ethnicity Face Anti-spoofing

We propose a statistical learning approach for constructing an evaluation function for face alignment. A nonlinear classification function is learned from a set of positive (good alignment) and negative (bad alignment) training examples to effectively distinguish between qualified and un-qualified alignment results. The AdaBoost learning algorithm is used, where weak classifiers are constructed based on edge features and combined into a strong classifier. Several strong classifiers are learned in stages using bootstrap samples during the training. The evaluation function thus learned gives a quantitative confidence and the good-bad classification is achieved by comparing the confidence with a learned optimal threshold. We point out the importance of using cascade strategy in the stagewise learning of strong classifiers. The divide-and-conquer strategy not only dramatically increases the speed of classification, but also makes the training easier and the good-bad classification more effective. Experimental results demonstrate that the classification function learned using the proposed approach provides semantically more meaningful scoring than the reconstruction error used in AAM for classification between qualified and un-qualified face alignment.

Stan Z. Li

Papers

DLME: Deep Local-Flatness Manifold Embedding

Single Shot Attention-Based Face Detector

Reducing Impact of Inaccurate User Feedback in Face Retrieval

Static and Dynamic Fusion for Multi-modal Cross-ethnicity Face Anti-spoofing

Evaluation of face alignment solutions using statistical learning