The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
ORB-SLAM is a feature-based simultaneous localization and mapping (SLAM) system. It has achieved good results in tracking, mapping and loop closing. However, the map created by ORB-SLAM with the monocular camera can not get the real scale. This paper presents an improving ORB SLAM system that helps to alleviate this issue by defining a baseline initialization procedure. We take two relative poses...
An improved KNN text classification algorithm based on Simhash has been proposed by introducing Simhash and the average Hamming distance of adjacent texts as a unit, which solves the problems caused by data imbalance and the large computational overhead in the traditional KNN text classification algorithms. Experimental results demonstrate that the proposed algorithm performs a higher precision, a...
In this paper we propose a deep learning architecture to make the best use of global and local information for pixel-wise semantic segmentation. The architecture of three-skips CNN is built with convolutional layers in VGG16 network and its mirrored convolutional layers. Our architecture aims to road scene understanding. In order to save memory and computational time, we use unpooling layers to map...
Traditional tomography methods are based on a single statistical feature(such as delay difference, variance, etc.) of network as a measure of the shared path length, but the single statistics don't reflect the time-varying of network performance parameters. In this paper, the packet group detection model is used to obtain the end-to-end delay curve in the non-stationary network for the first time...
As our population ages, neurological impairments and degeneration of the musculoskeletal system yield gait abnormalities, which can significantly reduce quality of life. Gait rehabilitative therapy has been widely adopted to help patients maximize community participation and living independence. To further improve the precision and efficiency of rehabilitative therapy, more objective methods need...
The deep learning based trackers can always achieve high tracking precision and strong adaptability in different scenarios. However, due to the fact that the number of the parameter is large and the fine-tuning is challenging, the time complexity is high. In order to improve the efficiency, we proposed a tracker based on fast deep learning through constructing a new network with less redundancy. Based...
This paper deals with image categorization from weak supervision, e.g. global image labels. We propose to improve the region selection performed in latent variable models such as Latent Support Vector Machine (LSVM) by leveraging human eye movement features collected from an eye-tracker device. We introduce a new model, Gaze Latent Support Vector Machine (G-LSVM), whose region selection during training...
Solar photovoltaic system maybe underperformanned because of too much dust, dirt and bird droppings, etc. 3D mapping for the cleaning robot still remains challenging in this large scale scenario. This paper presents a robust mapping system with Kinect V2 for automated photovoltaic cleaning system. Firstly, Kinect V2 is well-calibrated for a remedy to the production variety and the approach is fast...
In this paper we present a framework for improving face recognition system that have several stages. Some improvements of every stage are very important to the recognition results. Driven by this intuition, we proposed a novel scheme that gives the system a better performance. The scheme including dataset augment for learning, especially for big data requirement of deep learning. Enhancing the image...
With the advancement and development of the Internet, Flash has penetrated into all aspects of people's lives. As a popular multimedia, Flash has the advantages of strong artistic expressions, simple manufacture, flexible interaction and small storage, so Flash is widely used on the Internet. Currently, the shortages of researches about extracting content features of Flash seriously impose restrictions...
Multimedia data, especially image and video data, have become one of the most overwhelming data types on the Internet recently. Considering the user experience and real application requirements, multimedia data always demand a real-time processing speed. As a result, the huge amount of such data make retrieving useful information from them not only data-intensive, but also computation-intensive, which...
According to the image quality degradation in super-resolution reconstruction, we present a new algorithm for a single image super-resolution reconstruction to improve the image resolution. Considering the limitations that the extracted features of low-resolution image can not adapt to image direction changes, we propose a new feature-extracted method, and we build a new global optimization framework...
This paper deals with automatic systems for image recipe recognition. For this purpose, we compare and evaluate leading vision-based and text-based technologies on a new very large multimodal dataset (UPMC Food-101) containing about 100,000 recipes for a total of 101 food categories. Each item in this dataset is represented by one image plus textual information. We present deep experiments of recipe...
In recent years, mining micro-blog becomes a hot research field, especially it may create commercial and political values in a fast changing big data era. This paper investigates the sentiment analysis of Chinese micro-blogs (SACM) using a vector space model. With the analysis of the nature properties of the Chinese micro-blogs, a sentiment analysis system has been proposed by formulating it as a...
Spectrum sensing is a critical prerequisite in envisioned applications of wireless cognitive sensor networks which promise to resolve the perceived bandwidth scarcity versus under-utilization dilemma. The Kernel method is a very powerful tool in machine learning. The trick of kernel has been effectively and extensively applied in many areas of machine learning. In this paper, we propose a novel spectrum...
Aiming to the problem of weak primary user signal detection rate in low signal-to-noise ratio environments, we propose a novel spectrum sensing method based on the principal component analysis (PCA) and random forest (RF). From the received radio signal, a set of cyclic spectrum features are first calculated, and the PCA is applied to extract the most discriminate feature vector for classification...
Tongue diagnosis plays an important role in Traditional Chinese Medicine. However, due to its experience-based nature, traditional tongue diagnosis has to be implemented by specialists and there is few computerized tongue diagnosis method. In this paper, we proposed a novel tongue diagnosis method developed on the basis of Java language for Android platform. It can obtain tongue outline accurately...
In our previous work, we have presented a cross-stream dependency modeling method for hidden Markov model (HMM) based parametric speech synthesis. In this method, multi-space probability distribution (MSD) was adopted for F0 modeling and the voicing decision error influenced the accuracy of generated spectral features severely. Therefore, a cross-stream dependency modeling method using continuous...
This paper introduces the design of a real time vision-based motion synthesis system. the system requires user to wear the markers in a certain color. Based on that, several novel algorithms were used for feature detection and feature tracking under occlusion by estimating the velocity of missing features based on the prior, smoothness and fitness term. These algorithms ensured the accuracy and low...
In order to extract the gait sequence features better, this paper presents a method of feature extraction algorithm based on Fisher Discriminance. First, extract the profile silhouette by background cut off method, and then get the gait cycle by the change of their center of mass, at last convert the feature space. Then weight the gait characteristic matrix by Fisher Discriminance to make characteristics...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.