Sorry, you need to enable JavaScript to visit this website.

Heavy fog degrades the quality of road images in losing contrast and color fidelity, which may cause errors in stereo matching and road segmentation for advanced driver assistance systems (ADAS). Accident rates can be reduced if robust and efficient algorithms are applied for fog removal. Most studies have been conducted on this subject to date, but existing methods are not suitable for heavy haze conditions. Local darkness, under-estimation and over enhancement are always occurred after dehazing.

In this paper, we propose a novel and effective method for improving the accuracy of separating reflection components in a single image based on the dichromatic reflection model after calculating the diffuse reflection component by any existing method. Separating reflection components accurately is very important and useful in computer vision to detect highlight areas, which are often regarded as outliers, and to enhance image quality, especially texture, because we can control the intensity and apply a filter independently to each reflection component.

High-fidelity virtual content is essential for the creation of compelling and effective virtual reality (VR) experiences.

Hand segmentation is a fundamental technology in computer vision, which is used as a pre-processing step for various applications such as human-computer interaction (HCI), sign language translation, and medical systems. This demo provides interaction-free hand segmentation using Kinect camera. It is completely automatic, which can automatically segment foreground from the background and select the seed on the depth image for hand segmentation using Kinect camera.

In this paper, we propose a saliency prediction algorithm utilizing generative adversarial networks. The proposed system contains two parts: saliency network and adversarial networks. The saliency network is the basis for saliency prediction, which calculates an Euclidean cost function on the grayscale values between the predicted saliency map and the ground truth. In order to improve the accuracy of the algorithm, adversarial networks are subsequently utilized to extract the features of input data by coordinating the learning rates of the two sub-networks contained in the networks.

In this paper, we propose an accurate generative adversarial networks based saliency prediction model. Saliency network is an intact model to produce saliency maps. With the help of adversarial networks, feature extraction is more smooth and thorough. Moreover, the fully convolutional networks in saliency network facilitate the continuity and accuracy of pixel values in a saliency map. Compared with the six stateof-the-art methods, the proposed model has achieved highest accuracy. Besides, the performance of our model indicates that adversarial networks could be applied to more than classification. For future work, we will extend the algorithm to semi-supervised saliency prediction since DCGAN is a strong candidate for unsupervised learning.

One of the most critical missions of sonar is to capture deep-sea pictures to depict sea floor and various objects, and provide an immense understanding of biology and geology in deep sea. Due to the poor condition of underwater acoustic channel, the captured sonar images very possibly suffer from several typical types of distortions before finally reaching to users. Unfortunately, very limited efforts have been devoted to collecting meaningful sonar image databases and benchmark reliable objective quality predictors.