Search

Abstract Traffic sign detection systems constitute a key component in trending real-world applications, such as autonomous driving, and driver safety and assistance. This paper analyses the state-of-the-art of several object-detection systems (Faster R-CNN, R-FCN, SSD, and YOLO V2) combined with various feature extractors (Resnet V1 50, Resnet V1 101, Inception V2, Inception Resnet V2, Mobilenet V1, and Darknet-19) previously developed by their corresponding authors. We aim to explore the properties of these object-detection models which are modified and specifically adapted to the traffic sign detection problem domain by means of transfer learning. In particular, various publicly available object-detection models that were pre-trained on the Microsoft COCO dataset are fine-tuned on the German Traffic Sign Detection Benchmark dataset. The evaluation and comparison of these models include key metrics, such as the mean average precision (mAP), memory allocation, running time, number of floating point operations, number of parameters of the model, and the effect of traffic sign image sizes. Our findings show that Faster R-CNN Inception Resnet V2 obtains the best mAP, while R-FCN Resnet 101 strikes the best trade-off between accuracy and execution time. YOLO V2 and SSD Mobilenet merit a special mention, in that the former achieves competitive accuracy results and is the second fastest detector, while the latter, is the fastest and the lightest model in terms of memory consumption, making it an optimal choice for deployment in mobile and embedded devices. [ABSTRACT FROM AUTHOR]

Micro-expression is one of important clues for detecting lies. Its most outstanding characteristics include short duration and low intensity of movement. Therefore, video clips of high spatial-temporal resolution are much more desired than still images to provide sufficient details. On the other hand, owing to the difficulties to collect and encode micro-expression data, it is small sample size. In this paper, we use only 560 micro-expression video clips to evaluate the proposed network model: Transferring Long-term Convolutional Neural Network (TLCNN). TLCNN uses Deep CNN to extract features from each frame of micro-expression video clips, then feeds them to Long Short Term Memory (LSTM) which learn the temporal sequence information of micro-expression. Due to the small sample size of micro-expression data, TLCNN uses two steps of transfer learning: (1) transferring from expression data and (2) transferring from single frame of micro-expression video clips, which can be regarded as “big data”. Evaluation on 560 micro-expression video clips collected from three spontaneous databases is performed. The results show that the proposed TLCNN is better than some state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

In this work, we address the problem of face verification, namely determining whether a pair of face images belongs to the same or different subjects. Previous works often consider solving the problem of face verification in two steps: feature extraction and face recognition, resulting in a fragmented procedure. We argue that these techniques, although working well, fail to explicitly exploit a full end-to-end framework for face verification, which has received much attention and achieved significant improvements recently. In this paper, we propose a novel Joint Bayesian guided metric learning technique for dealing with the face verification task, which well integrates the above two steps of face verification into an end-to-end convolutional neural network (CNN) architecture. In the training stage, an initial neural network, which has the similar architecture with GoogLeNet CNN model, is firstly pre-trained by optimizing classification-based objective functions on the publicly available CASIA WebFace database. Based on constructed face pairs dataset from CASIA WebFace and LFW datasets, we then fine-tune the whole network parameters under the guide of the learned knowledge, which is obtained from the highly successful Joint Bayesian model. This guided learning procedure, which can also be seen as a metric learning technique, can further update network parameters for discriminating face pairs. In the testing process, the outputs by this unified network are discriminated with a threshold value to produce the ultimate prediction for the face verification task. Comprehensive evaluations over the LFW dataset well demonstrate the encouraging face verification performance of our proposed framework. [ABSTRACT FROM AUTHOR]

In this paper, a novel image operator is proposed to detect and locate text in scene images. To achieve a high recall of character detection, extremal regions are detected as character candidates. Two classifiers are trained to identify characters, and a recursive local search algorithm is proposed to extract characters that are wrongly identified by the classifiers. An efficient pruning method, which combines component trees and recognition results, is proposed to prune repeating components. A cascaded method combines text line entropy with a Convolutional Neural Network model. It is used to verify text candidates, which reduces the number of non-text regions. The proposed technique is test on three public datasets, i.e. ICDAR2011 dataset, ICDAR2013 dataset and ICDAR2015 dataset. The experimental results show that our approach achieves state-of-the-art performance. [ABSTRACT FROM AUTHOR]

Since the proposal of a fast learning algorithm for deep belief networks in 2006, the deep learning techniques have drawn ever-increasing research interests because of their inherent capability of overcoming the drawback of traditional algorithms dependent on hand-designed features. Deep learning approaches have also been found to be suitable for big data analysis with successful applications to computer vision, pattern recognition, speech recognition, natural language processing, and recommendation systems. In this paper, we discuss some widely-used deep learning architectures and their practical applications. An up-to-date overview is provided on four deep learning architectures, namely, autoencoder, convolutional neural network, deep belief network, and restricted Boltzmann machine. Different types of deep neural networks are surveyed and recent progresses are summarized. Applications of deep learning techniques on some selected areas (speech recognition, pattern recognition and computer vision) are highlighted. A list of future research topics are finally given with clear justifications. [ABSTRACT FROM AUTHOR]

This paper focuses on the enhancement of the generalization ability and training stability of deep neural networks (DNNs). New activation functions that we call bounded rectified linear unit (ReLU), bounded leaky ReLU, and bounded bi-firing are proposed. These activation functions are defined based on the desired properties of the universal approximation theorem (UAT). An additional work on providing a new set of coefficient values for the scaled hyperbolic tangent function is also presented. These works result in improved classification performances and training stability in DNNs. Experimental works using the multilayer perceptron (MLP) and convolutional neural network (CNN) models have shown that the proposed activation functions outperforms their respective original forms in regards to the classification accuracies and numerical stability. Tests on MNIST, mnist - rot - bg - img handwritten digit, and AR Purdue face databases show that significant improvements of 17.31%, 9.19%, and 74.99% can be achieved in terms of the testing misclassification error rates (MCRs), applying both mean squared error (MSE) and cross-entropy (CE) loss functions This is done without sacrificing the computational efficiency. With the MNIST dataset, bounding the output of an activation function results in a 78.58% reduction in numerical instability, and with the mnist-rot-bg-img and AR Purdue databases the problem is completely eliminated. Thus, this work has demonstrated the significance of bounding an activation function in helping to alleviate the training instability problem when training a DNN model (particularly CNN). [ABSTRACT FROM AUTHOR]

Action recognition (AR) is one of the most important tasks in video analysis and computer vision. Recently a large number of related methods have been proposed. While most of these methods are investigated on AR datasets collected from the visible spectrum, the AR problem under infrared scenarios still has not attracted much attention. There is even few public infrared datasets available for supporting the fundamental evaluation requirements of this research. To this issue, this work aims to emphasize the importance of the infrared AR problem in applications and arouse researchers' attention on this task. Specifically, we construct a new Inf rared A ction R ecognition ( InfAR ) dataset captured at different times, including in summer and winter, and explore how discriminable actions in our InfAR dataset are with the state-of-the-art pipelines based on low-level features and deep convolutional neural network (CNN), respectively. Our results reveal: (1) In all, dense trajectory feature can achieve the best performance while the appearance features, e.g., HOG, have relatively poorer performance; (2) the encoding method of vector of locally aggregated descriptors is evidently better than that of the widely-used Fisher Vector; (3) the late fusion facilitates a better performance than early fusion; (4) action videos captured in winter is more discriminable than in summer; (5) compared to appearance information, the motion information is more essential for infrared action recognition and utilizing this information through deep CNN can improve greatly the performance. The best performance achieved on our dataset is 76.66% (Average Precision), leaving a reasonable space for further exploring the insights underlying such type of infrared AR problem and accordingly designing proper techniques to further promote the performance on this specifically constructed InfAR dataset. [ABSTRACT FROM AUTHOR]