S
Saeed Ranjbar Alvar
Researcher at Simon Fraser University
Publications - 26
Citations - 190
Saeed Ranjbar Alvar is an academic researcher from Simon Fraser University. The author has contributed to research in topics: Collaborative intelligence & Cloud computing. The author has an hindex of 6, co-authored 22 publications receiving 128 citations. Previous affiliations of Saeed Ranjbar Alvar include University of Tabriz & Middle East Technical University.
Papers
More filters
Proceedings ArticleDOI
Multi-Task Learning with Compressible Features for Collaborative Intelligence
TL;DR: In this article, a new loss function was proposed to encourage feature compressibility while improving system performance on multiple tasks, which can achieve around 20% bitrate reduction without sacrificing the performance on several vision-related tasks.
Proceedings ArticleDOI
MV-YOLO: Motion Vector-Aided Tracking by Semantic Object Detection
TL;DR: A hybrid tracker that leverages motion information from the compressed video stream and a general-purpose semantic object detector acting on decoded frames to construct a fast and efficient tracking engine is presented.
Proceedings ArticleDOI
Bit Allocation for Multi-Task Collaborative Intelligence
TL;DR: This paper establishes a model for the joint distortion of the multiple tasks as a function of the bit rates assigned to different deep feature tensors, and solves the rate-distortion optimization problem under a total rate constraint to obtain the best rate allocation among the tensors to be transferred.
Journal ArticleDOI
Pareto-Optimal Bit Allocation for Collaborative Intelligence.
TL;DR: This article model task distortion as a function of rate using convex surfaces similar to those found in distortion-rate theory and provides analytical characterization of the full Pareto set for 2-stream $k$ -task systems, and bounds on the Pare to set for 3-stream 2-task systems.
DFTS: Deep Feature Transmission Simulator
TL;DR: This work presents a simulator to help study the effects of imperfect packet-based transmission of deep features over an imperfect channel, implemented in Keras and allows users to study theeffects of both lossy packet transmission and quantization on the accuracy.