Sharif Digital Repository

In this paper we have proposed a multi-agent architecture for an Intelligent Tutoring System using TROPOS methodology. Basic assumptions of constructivist learning have been taken into account in analysis and design activities. Nine types of agents are distinguished in the presented system. We have centralized management in the proposed architecture which is done by an agent called Manager Agent. This strategy of centralized management will result in a more coherent and effective Tutoring System than other systems not using such an agent, since it applies control flow of agents and decreases agents ' useless communications which may cause destructive congestion in system. This text discusses... 

This paper presents a novel video encoding method for cooperative Educational Dissemination Systems. Taking into consideration the inherent characteristics of distance learning video streams, existing a few moving objects in the scene and objects having slow motions, we propose a novel content-based video encoding method which is very efficient on low bandwidth channels. In the encoding process, we apply a background subtraction algorithm for motion segmentation with a novel statistical background modeling. In each frame, the moving objects are extrapolated with rectangular bounding boxes which are the only data send over the low bandwidth channel. In the decoding process, we propose a new... 

This paper presents a novel method for cooperative educational dissemination systems. Taking into consideration the inherent characteristics of distance learning video streams (existence of a few slow moving objects in a classroom), we have proposed a novel content-based video coding method that is very efficient at low bitrate channels. On the encoding side, we have applied a background subtraction algorithm for motion segmentation using a novel statistical background modeling approach. At each frame, the moving objects are extrapolated with a rectangular model and tracked frame by frame (which forms the only data needed to be sent over the channel). On the decoding side, we have used a new... 

This paper presents a novel video encoding method for cooperative educational dissemination systems. Taking into consideration the inherent characteristics of stereo cameras framework in our educational videos and the ability of determining objects in different depths in a scene, we have proposed a novel object-based video encoding based on "sprite coding" that supports the MPEG-4 Version 1 Main profile in order to transfer distance learning videos across narrow-band transmission links such as the Internet. This paper proposes a multi-layer video object layer generation scheme with foreground moving object extraction and background sprite generation using stereo camera property. The... 

Federated learning (FL) and split learning (SL) are state-of-the-art distributed machine learning techniques to enable machine learning without accessing raw data on clients or end devices. However, their comparative training performance under real-world resource-restricted Internet of Things (IoT) device settings, e.g., Raspberry Pi, remains barely studied, which, to our knowledge, have not yet been evaluated and compared, rendering inconvenient reference for practitioner. This work firstly provides empirical comparisons of FL and SL in real-world IoT settings regarding learning performance and on-device execution overhead. Our analyses demonstrate that the learning performance of SL is... 

This paper presents a novel and fast homography-based tracking method to segment and track the instructor placed in front of a class board using two synchronized views. In the proposed method we estimate the instructor as a planar object and track its planar parameters. At first, the homography of board and fundamental geometry of two views are extracted using six point correspondences. The homography transformation matrix of object is calculated by obtaining three point correspondences in parallax region introduced with homography of the board. The parameters to be tracked by Kalman filter are obtained frame by frame by object's homography decomposition. Our proposed method is robust...