Correspondence selection, a crucial step in many computer vision tasks, aims to distinguish between inliers and outliers from putative correspondences. The coherence of correspondences is often used for predicting inlier probability, but it is difficult for neural networks to extract coherence contexts based only on quadruple coordinates. To overcome this difficulty, we propose enhancing the preliminary features using local and global handcrafted coherent characteristics before model learning, which strengthens the discrimination of each correspondence and guides the model to prune obvious outliers. Furthermore, to fully utilize local information, neighbors are searched in coordinate space as well as feature space. These two kinds of neighbors provide complementary and plentiful contexts for inlier probability prediction. Finally, a novel neighbor representation and a fusion architecture are proposed to retain detailed features. Experiments demonstrate that our method achieves state-of-the-art performance on relative camera pose estimation and correspondence selection metrics on the outdoor YFCC100M [1] and the indoor SUN3D [2] datasets.

Local feature detection and description are two essential steps in many visual applications. Most learned local feature methods require high-quality labeled data to achieve superior performance, but such labels are often expensive. To address this problem, we propose MUFeat, an unsupervised learning framework of jointly learning local feature detector and descriptor without requirement of ground-truth correspondences. MUFeat trains the network based on the putative matches from the pretrained model and two proposed unsupervised loss functions. Furthermore, the MUFeat framework includes a pyramidal feature hierarchy network to obtain keypoints and descriptors from feature maps. Experiments indicate that MUFeat outperforms most state-of-the-art supervised learning methods on image matching, medical image registration and visual localization tasks.

Traffic in wireless networks has become more diverse due to various applications running on smartphones. Moreover, traffic generated by background processes which has small amount of data and widely dispersed in time takes up a large portion of diverse data applications. To reduce signaling overhead due to frequent state transitions of radio resource control (RRC), the evolved nodeB (eNB) tends to keep User equipment (UE) in RRC connected state all the time. However, this approach leads to high UE power consumption and inefficient resource utilization at eNB, especially when UE has only background traffic. To cope with this problem, we propose three shaping schemes which transmit together accumulated background data to reduce both signaling overhead and power consumption at the same time. In addition, we analyze the performance of the proposed shaping schemes. The analytical results are verified with simulations.

Energy saving and quality of service (QoS) support are two important issues of Long-Term Evolution Advanced (LTE-A) wireless networks. Unfortunately, achieving high energy saving efficiency often implies degradation of QoS support. In this paper, we address the issue of QoS support under Discontinuous Reception (DRX), the power saving mechanism of LTE-A. Different from previous studies which analyzed average packet delay under DRX mechanism, we adopt packet loss ratio due to violation of delay bound as QoS requirements of real-time applications. The problem of QoS support under DRX mechanism is defined as an optimization problem to maximize average power saving under the constraint that the packet loss ratio is no greater than a pre-determined threshold. We derive packet loss ratio and energy saving efficiency given DRX parameter values and propose an approach to select appropriate DRX parameter values. Simulations are conducted to verify our analysis. We found that the analytical results match well with simulations. Moreover, the energy saving efficiency achieved by the selected DRX parameter values is very close to that of the optimum solution.