In this paper, we propose a novel multilayer sparse coding network capable of efficiently adapting its own regularization parameters to a given dataset. The network is trained end-to-end with a supervised task-driven learning algorithm via error backpropagation. During training, the network learns both the dictionaries and the regularization parameters of each sparse coding layer so that the re...

Part of the information obtained by rodent whiskers is carried by the frequency of their movement. In the thalamus of anesthetized rats, the whisker frequency is represented by two different coding schemes: by amplitude and spike count (i.e., response amplitudes and spike counts decrease as a function of frequency) in the lemniscal thalamus and by latency and spike count (latencies increase and...

In this paper, we propose a novel scalable video coding algorithm based on multiple reference frame method. To improve the coding efficinecy in both base layer and enhancement layer, double reference picture and double motion vector algorithms are employed for motion compensated prediction. Moreover, these two double reference picture and double motion vectot schemes can reduce drift problem, w...

Fine-grained scalable (FGS) coding of video streams has been proposed in the literature to accommodate client heterogeneity. FGS streams are composed of two layers: base layer, which provides basic quality, and a single enhancement layer that adds incremental quality refinements proportional to number of bits received. The base layer uses nonscalable coding which is more efficient in terms of c...

In this paper, we propose a new algorithm, which utilizes the enhancement layer prediction to further improve the coding efficiency of current Fine Granularity Scalability (FGS) defined in MPEG-4. The proposed algorithm adaptively uses (1) the previously reconstructed enhancement layer macroblock after motion compensation along with (2) current reconstructed base layer macroblock and (3) the co...

The JVT/H.26L video coding scheme conceptually consists of a Video Coding Layer (VCL) responsible mainly for coding efficiency and a Network Abstraction Layer (NAL) that supports video specific transport features for a variety of networks. This paper describes the H.26L/JVT NAL in general, including the parameter set concept and the transport over packet-based and bit-stream oriented networks. ...

The Scalable Video Coding (SVC) has been recently added as an extension to H.264/AVC standard. This extension allows both bit rate and device capability adaptation which are desirable features especially in error-prone wireless heterogeneous networks. This paper investigates the spatiotemporal error concealment techniques for packet losses in wireless IP networks, where scalable video coding (S...

Providing cross-layer quality of service (QoS) for wireless multimedia delivery is challenging task which has attracted intensive research over the years. A fundamental problem in this area is how to map QoS criterion at different layers and optimum QoS across the layers. We present end-to-end QoS architecture for multimedia delivery as well as QoS metrics over wireless links together with chan...

In Multiple Input Multiple Output (MIMO) technology, a cross layer design enhances the spectral efficiency, reliability and throughput of the network. In this paper, a cross-layer approach for k-NN based Adaptive Modulation Coding (AMC) and Incremental Redundancy Hybrid Automatic Repeat Request (IR-HARQ) in MIMO is proposed. The proposed cross layer approach connects physical layer and data lin...