Search

This paper focuses on designing a medium access control and channel usage algorithm in multi-rate wireless local area networks for improving the efficiency and fairly sharing channel resources among the contending nodes. Aiming for the problems that the high collision is often caused by binary exponential backoff mechanism in the legacy IEEE 802.11 and the shared channel can be overused by low bitrate nodes, we propose a differentiated reservation (DR) algorithm to reduce the collision among the contending nodes by setting their backoff counter as a deterministic value once accessing successfully to the channel. Furthermore, to eliminate the performance anomaly, some nodes are permitted to send multiple packets in one transmission opportunity according to their feature. Moreover, we present the implementation of the DR algorithm that is readily applied to both the existing 802.11 DCF and 802.11e EDCA networks with minimum modification. In addition, we also investigate the limitation of the DR algorithm and propose a group-based differentiated reservation (GDR) algorithm applied to high dense scenarios. The results of the theoretical analysis and simulation validate that our proposed algorithms (DR and GDR) can obtain high throughput, good airtime fairness, and low collision rate.

With the development of deep learning, image super-resolution has made great breakthroughs. However, compared with a color image, the performance of depth map super-resolution is still poor. To address this problem, multilevel recursive guidance and progressive supervised network (MRG-PS) is proposed in this paper. First, a multilevel recursive guidance architecture is presented to extract features of a color stream and depth stream, in which the depth stream is guided by the color features at each level. Second, a progressive supervision module is developed to supervise the multilevel recursion to obtain depth residual information on different levels. Finally, a residual fusion and construction strategy is designed to fuse all residual information and reconstruct the high-resolution depth map. The experimental results demonstrate that the proposed method outperforms the state-of-the-art methods.

Sketch-based image retrieval (SBIR) is a challenging task due to the cross-domain gap between sketch queries and natural images. In this paper, we propose a novel edge-guided cross-domain learning network to reduce the domain gap. In particular, edge maps extracted from natural images are introduced as the bridge between two domains, and the inputs from different domains are embedded into a common feature space. An edge guidance module is proposed to fuse natural images and the corresponding edge maps, which guides the network to generate more discriminative features in the domain alignment process. Meanwhile, a shape regression module is proposed to capture the inherent shape similarity between sketches and natural images. By training the proposed network in an end-to-end process, the sketch and natural image domains can be effectively associated, which potentially overcomes the challenge of the common feature learning for two heterogeneous domains. The experimental results on the SBIR dataset demonstrate that the proposed method achieves superior performance compared with the state-of-the-art methods.

With the development of 3-D applications, such as 3-D reconstruction and object recognition, accurate and high-quality depth map is urgently required. Recently, depth cameras have been affordable and widely used in daily life. However, the captured depth map always owns low resolution and poor quality, which limits its practical application. This paper proposes a color-guided depth map super resolution method using convolutional neural network. First, a dual-stream convolutional neural network, which integrates the color and depth information simultaneously, is proposed for depth map super resolution. Then, the optimized edge map generated by the high resolution color image and low resolution depth map is used as additional information to refine the object boundary in the depth map. Experimental results demonstrate the effectiveness of the proposed method compared with the state-of-the-art methods.