Search

Three-dimensional object detection from point cloud data is becoming more and more significant, especially for autonomous driving applications. However, it is difficult for lidar to obtain the complete structure of an object in a real scene due to its scanning characteristics. Although the existing methods have made great progress, most of them ignore the prior information of object structure, such as symmetry. So, in this paper, we use the symmetry of the object to complete the missing part in the point cloud and then detect it. Specifically, we propose a two-stage detection framework. In the first stage, we adopt an encoder&ndash, decoder structure to generate the symmetry points of the foreground points and make the symmetry points and the non-empty voxel centers form an enhanced point cloud. In the second stage, the enhanced point cloud is input into the baseline, which is an anchor-based region proposal network, to generate the detection results. Extensive experiments on the challenging KITTI benchmark show the effectiveness of our method, which has better performance on both 3D and BEV (bird&rsquo, s eye view) object detection compared with some previous state-of-the-art methods.

In this paper, we propose a novel reticular shield detection algorithm to restore the images that are obscured by reticular shield such as barbed wires and fences. The whole framework is composed of three stages: stage-1 for accurate superpixel segmentation, stage-2 for deriving the mask of reticular shield and stage-3 for restoring the image by introducing the algorithm. In this paper, to obtain the accurate mask of the shield, a new joint feature CCTP is used to train support vector machine (SVM) classifier and classify all superpixels. Finally, we make comparison of the effects of the proposed algorithm and several other algorithms put forward by other scientists. We introduce two indexes, PSNR and SSIM to evaluate the results and the proposed algorithm is obviously superior to the other two algorithms, which can also be proved by looking into the restoration images.

Multimedia services have increased sharply in recent years, especially in the video domain. This trend raises the demand for video quality assessment. The main factors affecting the video quality are compression technology and transmission link. H.265/High Efficiency Video Coding (HEVC), which serves as the latest video coding standard, has shown significant improvements on the video compression when compared with the last video coding standard H.264/MPEG-4 AVC. This paper evaluates the video quality under H.265/HEVC compression standard and studies the relationship between the video quality and the bitrate under different group of pictures (GOP) patterns. The evaluation is done for four video sequences using objective metrics. Experiments demonstrate that higher bitrate will lead to better video quality under the same GOP pattern. Also, the video quality will improve as the size of GOP increases while maintaining the number of B-frames. Meanwhile, the effectiveness of the compression also depends on the content and dynamic condition of the video sequences.