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

Convolutional neural networks (CNNs) have been widely used in computer vision community, significantly improving the state-of-the-art. In most of the available CNNs, the softmax loss function is used as the supervision signal to train the deep model. In order to enhance the discriminative power of the deeply learned features, this paper proposes a new supervision signal, called center loss, for face recognition task. Specifically, the center loss simultaneously learns a center for deep features of each class and penalizes the distances between the deep features and their corresponding class centers. More importantly, we prove that the proposed center loss function is trainable and easy to optimize in the CNNs. With the joint supervision of softmax loss and center loss, we can train a robust CNNs to obtain the deep features with the two key learning objectives, inter-class dispension and intra-class compactness as much as possible, which are very essential to face recognition. It is encouraging to see that our CNNs (with such joint supervision) achieve the state-of-the-art accuracy on several important face recognition benchmarks, Labeled Faces in the Wild (LFW), YouTube Faces (YTF), and MegaFace Challenge. Especially, our new approach achieves the best results on MegaFace (the largest public domain face benchmark) under the protocol of small training set (contains under 500000 images and under 20000 persons), significantly improving the previous results and setting new state-of-the-art for both face recognition and face verification tasks.

/pdf/a-discriminative-feature-learning-approach-for-deep-face-2zrw7nm451.pdf

A Discriminative Feature Learning Approach for Deep Face Recognition

The PASCAL Visual Object Classes Challenge

This paper addresses deep face recognition (FR) problem under open-set protocol, where ideal face features are expected to have smaller maximal intra-class distance than minimal inter-class distance under a suitably chosen metric space. However, few existing algorithms can effectively achieve this criterion. To this end, we propose the angular softmax (A-Softmax) loss that enables convolutional neural networks (CNNs) to learn angularly discriminative features. Geometrically, A-Softmax loss can be viewed as imposing discriminative constraints on a hypersphere manifold, which intrinsically matches the prior that faces also lie on a manifold. Moreover, the size of angular margin can be quantitatively adjusted by a parameter m. We further derive specific m to approximate the ideal feature criterion. Extensive analysis and experiments on Labeled Face in the Wild (LFW), Youtube Faces (YTF) and MegaFace Challenge 1 show the superiority of A-Softmax loss in FR tasks.

/pdf/sphereface-deep-hypersphere-embedding-for-face-recognition-bof1bk9snh.pdf

SphereFace: Deep Hypersphere Embedding for Face Recognition

A scheme is developed for classifying the types of motion perceived by a humanlike robot. It is assumed that the robot receives visual images of the scene using a perspective system model. Equations, theorems, concepts, clues, etc., relating the objects, their positions, and their motion to their images on the focal plane are presented. >

http://ieeexplore.ieee.org/iel2/815/3148/00100651.pdf

Robot vision

Video transcoding is a popular technique for adapting the bit-rate or spatial/temporal resolution of a precoded video to suit better the constraints and requirements of different transmission networks and receiving devices. To minimize computational complexity, many fast methods have been proposed to obtain the motion vectors required for transcoding a precoded video through reducing its frame size by an integral factor. In this paper, we extend the existing work by developing and comparing several fast methods of downsizing precoded videos by a rational factor. Methods that outperform others under different conditions or with different computational requirements are identified, and an application scenario that can benefit from the proposed rational downsizing video transcoding is presented. An efficient scheme is also proposed to select the proper reduced frame size for sustaining the best possible video quality at a specified lower bit-rate. The superiority of the proposed transcoding approach in comparison with the existing integral downsizing video transcoding or cascaded video re-encoding methods is evident from the experimental results shown in this paper.

On the methods and performances of rational downsizing video transcoding

This invention describes an audio-visual based method and system for early drowning detection system. In this invention, a number of cameras (100, 200, 201) are mounted on top of a swimming pool (101). These cameras (100, 200, 201) are used to monitor swimmers in the pool (101) together with the aid of an array of microphones (103). Similarly, the microphone array (103) is mounted above the water to cover the entire swimming pool (101). Based on the motion and activity of the swimmer detected through the video camera (100, 200, 201), the swimmer's condition is automatically analyzed. Such automated video analysis includes building the visual background model of the pool (101), detecting the presence of the swimmers in the monitored areas (106), estimating the number of swimmers inside the monitored area (106), tracking the swimmers and analyzing the behaviour of each tracked swimmer in terms of body orientation, moving direction and motion patterns. In addition, the microphone array (103) is deployed to pick up audio signals (104) originating from distress calls. Once the system detects the presence of a potential drowning, both visual and audio alarms will be activated to draw the attention of the person in charge for further confirmation and if necessary to provide necessary follow-up rescue actions.

Drowning early warning system

In this paper, we propose a novel approach for scalable video delivery over multiuser Multiple Input Multiple Output-Orthogonal Frequency Division Multiplexing (MIMO-OFDM) systems with MAXMIN fairness ensured. Scalable Video Coding (SVC) is efficient for rate adaptation and unequal protection in wireless networks where heterogeneous mobile clients have different hardware specifications and time-varying channel conditions. Using the scalability characteristics of SVC, we propose a general approach to ensure fairness for multiple scalable video downlink from a multiple-antenna base station. The fairness is achieved by packet priority analysis referring to layer dependency from application layer and a MAXMIN resource allocation at the physical layer. Based on packet priorities, time-frequency resource, power and modulation schemes are adaptively selected for transmission. This scheme significantly improves the overall system performance and guarantee fairness among users, as demonstrated by experimental comparisons with conventional radio resource allocation schemes.

A MAXMIN resource allocation approach for scalable video delivery over multiuser MIMO-OFDM systems

Unsupervised clustering of dominant scenes in sports video

We propose a shot boundary detection approach based on unsupervised scenelet clustering and hypothesis testing. We define a video scenelet as a short consecutive samples of frames of a video sequence. The approach makes use of a typical k-means clustering algorithm to group the scenelets. Based on the clustering result, hypothesis testing can be performed to identify the shot boundaries at each level with a different cluster number. Combined with a cluster validity analysis to decide a suitable number of clusters, promising results can be obtained for shot boundary detection.

http://ieeexplore.ieee.org/iel5/9319/29625/01346333.pdf

Yap-Peng Tan

Papers

On the methods and performances of rational downsizing video transcoding

Drowning early warning system

A MAXMIN resource allocation approach for scalable video delivery over multiuser MIMO-OFDM systems

Unsupervised clustering of dominant scenes in sports video

Shot boundary detection using unsupervised clustering and hypothesis testing