Sharif Digital Repository

In this paper, we propose an information theoretic approach to fuse images compressed by compressed sensing (CS) techniques. The goal is to fuse multiple compressed images directly using measurements and reconstruct the final image only once. Since the reconstruction is the most expensive step, it would be a more economic method than separate reconstruction of each image. The proposed scheme is based on calculating the result using weighted average on the measurements of the inputs, where weights are calculated by information theoretic functions. The simulation results show that the final images produced by our method have higher quality than those produced by traditional methods, especially... 

In this paper, a solution is proposed to render both rough and detailed images information using only audio-range sound. The procedure is implemented in consecutive stages with stage-dependent parameters selected by the user. The first stage consists of edge detection and tracking of the boundaries of the objects to obtain the sketch of the image. In the second stage, the user can selectively access the details of the image, such as brightness, texture and spatial position of the objects. Practical methods are proposed and tested which accelerate the training process. As a particular application, an educational module is presented which assists students of a special school for the blind by... 

Auto-Exposure detection is a key feature in digital cameras and smart-phones. The image information can be used to detect scene over and under exposure passively. This paper introduces an effective optimal exposure detection function based on joint probability density function of the pixels in the scene  

Transformer networks, originally developed in the community of machine translation to eliminate sequential nature of recurrent neural networks, have shown impressive results in other natural language processing and machine vision tasks. Self-attention is the core module behind visual transformers which globally mixes the image information. This module drastically reduces the intrinsic inductive bias imposed by CNNs, such as locality, while encountering insufficient robustness against some adversarial attacks. In this paper we introduce K-sparse attention to preserve low inductive bias, while robustifying transformers against adversarial attacks. We show that standard transformers attend... 

Background: Transcranial direct current stimulation (tDCS) is a noninvasive brain stimulation tool suited to alter cortical excitability and activity via the application of weak direct electrical currents. An increasing number of studies in the addiction literature suggests that tDCS modulates subjective self-reported craving through stimulation of dorsolateral prefrontal cortex (DLPFC). The major goal of this study was to explore effects of bilateral DLPFC stimulation on resting state networks (RSNs) in association with drug craving modulation. We targeted three large-scale RSNs; the default mode network (DMN), the executive control network (ECN), and the salience network (SN). Methods:...