Transfer learning is a deep learning technique has proved to be of great importance. However, most of the standard transfer learning algorithms are designed to repeat the same method for fine-tuning of the weights on the target domain. If we try to ...

The main goal of this paper is to forecast the sales of airline tickets in time series affected by many factors including different flights, date features and short, middle, long-term historical sales information. The Deep Neural Network (DNN) model and ...

Generating a sequence of images from a multi-sentence paragraph is a recently proposed task called Story-Visualization. In this task, how to keep the global consistency across dynamic scenes and characters in the story flow is the distinct difference ...

This paper presents an autonomous method to realize the annotation of LiDAR (Light Detection And Ranging) point clouds with the help of images. Different from directly labeling on point clouds, our approach first utilizes mature image segmentation ...

Recognizing text from images taken in natural scenes is a challenging task and a hot research topic in computer vision. Unlike traditional optical character recognition (OCR), words in natural images often possess irregular layout (e.g. arbitrarily ...

Optimal features selection is essential for sentence similarity recognition. In this paper, a semantic matching model based on optimal selection mechanism is designed. Compared with other models that without using an optimal feature selection mechanism, ...

With the development of the shipping industry, a large number of long-term ships on the sea have serious problems in the accumulation of hull attachments. It is difficult and inefficient to clean up the attachment manually. It is one of the main ways to ...

With the development of neural network, the demand of storage and memory bandwidth is increasing rapidly. At the same time, most of the data in the neural network can be reduced and compressed. This paper analyzes various techniques of neural network ...

This paper investigates the problem of generating images from visual attributes and vice versa. Given the prevailing research of image recognition, the bi-directional generation between attributes and images is rarely explored due to the challenges of ...

Brain is the most advanced part of the nervous system. As an emerging neighborhood exploring the brain, the brain-computer interface (BCI) may completely influence people's current communication and lifestyle in the future. We propose an architecture ...

In this paper, we propose a method for synthesis of datasets with specific characteristics for the clustering task. Namely, we propose an algorithm, which can generate a clustering dataset given its meta-feature description. The method we propose is ...

With the raise of big data, machine learning and crowdsourcing, the volume of existing datasets for different machine learning problems have greatly increased. The natural language processing field is not an exception; so, as a result, most of the ...

This paper proposes an improved deep residual neural network for the classification of breast cancer as either benign or malignant. Inspired by the success of using focal loss in object detection, we present a new focal loss for pathological image ...

With the advent of the era of big data, it is necessary to obtain volume, veracity information on fishing activities. In this paper, we present a new approach to predict fishing activities using DeepFM with historical fishing data of Chinese vessels. ...

In large-scale traffic video analysis, continuous tracking of vehicles across cameras overcomes the time and space limitations of a single camera, and is conducive to transportation design and traffic flow optimization. In this work, we propose an end-...

In recent years, object detection technology has been continuously developed, and the tracking-by-detection strategy has gradually become the main method of multi-object tracking. Based on detection, the accuracy of the multi-object tracking depends on ...

The existing rating system, as the foundation of estimating results of competitive sports and games, has problems of insufficient use of competing information and the unilateral considering of competing process modeling. In this paper, the Artificial ...

The concept of ultrasound therapy has been proposed long ago. However, most of the previous methods of ultrasound therapy destroy tissue through thermal effects and cause great damage to patients. In recent years, ultrasound cavitation therapy has ...

Convolutional neural network based on machine learning has become one of the most popular methods of image recognition and segmentation, but it needs huge data samples to get better results. In this study, based on U-Net network, a two-stage convolution ...

Using the characteristic that image patches can be represented sparsely on a selected over-complete dictionary, the sparse coding based super-resolution (ScSR) method pioneers a new avenue for the SR technology. However, ScSR will result in obvious ...