Hossein Hashemi
Bio
My name is "Hossein Hashemi". I am currently a student of Master of science in Engineering in Artificial Intelligence and Robotics at the Sapienza Università di Roma . My special interest is computer vision and machine learning but some theoretical computer science issues and Web developing are also of great appeal to me.
Please find more about my academic activities here or in my CV
Publications
Controlling the Depth of Anesthesia During Operation by Vital Parameters of the Patient’s Body Using a Fuzzy Controller,
  Mohammad Reza Yousefi, Nazanin Dehghani Samani, Hossein Hashemi, Fatemeh Shahlaei,
2018 IEEE International Symposium on Medical Measurements and Applications (MeMeA), August 2018.
Evaluating Feature Extractors and Dimension Reduction
Methods for near Infrared Face Recognition Systems,
  Sajad Farokhi, Siti Mariyam Shamsuddin, U.U Sheikh, Jan Flusser, Hossein Hashemi,
Jurnal Teknologi, 70 (2014),
23-33 (Scopus Indexed, eISNN=2180-3722).
Automatic Image Processing of Stained Tissue Cells Using Clustering And an
Improved Watershed Method
  Hossein Hashemi , Sahar Farajzadeh, Nadia
Bagheri,
In preparation.
Books:
  24 Stories From Newcomers Mazandaranis Workshop Members Of Hamnegar Sari.
,
ISBN: 9786001000829.
Patent:
  Commute Controlling In Vehicle (Intelligent Traffic Lights), Patent Publication
Number: 53864 ,Iran, (Oct/19/2008).
,
Academic Activities
A novel 3D human pose estimation algorithm from a single image
Having a three-dimensional reconstruction of the human body is an important issue for machine vision and extensively can be used in human-computer interaction, virtual reality, video surveillance, and video games topics.
while 2D human body pose estimation in few last years have prospered very well, recently considerable attention has been paid to 3D human pose estimation problems.
This project improves an algorithm that do 3D motion capture (MOCAP) with 2D image or video as the only input of process.
Human activities recognition by multivariate time series classification
Recognizing human movement and activities from video sequences or images is a challenging task for several years which have various domains and many applications including video surveillance systems, human-computer interaction, and robotics for human behavior characterization, require a system to understand the what human movement and what he's doing action.
In this work, will propose a new method to analyze the human body joints in every frame of video for classifying action and daily human activities based Reservoir computing approaches of multivariate time series classification.
The benefits of this research can be useful for ‘complex body posture recognition’ or some social action recognition.
(How find the kiss scenes in the titanic film?)
My visualization result of the right shoulder in time series.
Read human body language is one important thing on human-computer interaction
Nowadays the trend of human life is going to be so automotive using robots and artificial intelligence in all aspects of the industry. What is so interesting in modern life, is preparing a comprehensive and effective interaction between robots and human.
Prediction the human posture recognition is a novel and interesting topic for the pioneer and innovative industries like autonomous car, robot cameraman, video content understanding, Human-robot interaction.
Assessment the human behavior by artificial intelligence and decide according to this investigation, is what affect the result and reaction by robots and machines.
However there are a lot of methods and algorithm to control the reaction and response by the robots, but they are not effective to access the reality in the acceptable time together with high speed and also they are not capable to consider and investigate the reality in three-dimensional.
Expand the knowledge and improve this technology to have a correct and precise reaction of the machines will be the purpose of this project.
The advantage of result of this project is be useful for the autonomous cars when it faces in moving object, prediction the next behavior of that object which can be human and deciding this prediction correctly what we are looking for, or in the field of the robot cameraman it should be capable to predict precisely and follow the moving object.
Eye Gaze Tracking Using Kinect
Most commercial eye gaze tracking systems are based on the use of infrared lights. However, such systems may not work outdoor or may have a very limited head box for them to work. This project proposes a non-infrared based approach to track one's eye gaze with an RGBD camera (in our case, Kinect). The proposed method adopts a personalized 3D face model constructed off-line. To detect the eye gaze, our system tracks the iris center and a set of 2D facial landmarks whose 3D locations are provided by the RGBD camera. A simple onetime calibration procedure is used to obtain the parameters of the personalized eye gaze model. We compare the performance of the proposed method against the 2D approach using only RGB input on the same images, and find that the use of depth information directly from Kinect achieves more accurate tracking. As expected, the results from the proposed method are not as accurate as the ones from infrared-based approaches. However, this method has the potential for practical use with upcoming better and cheaper depth cameras.
My robotics experience
Personal Activities
Traveling
I love traveling. Although I am mostly busy with my research projects and the university courses, I take advantage of every opportunity to see new places. Exploring different cultures and lifestyles around the world has always been fascinating for me.
anyway, already I've hitchhiked more than 7000-kilometers in my life.
Mountain Climbing
I love going mountains and spend my free time with my friends there. I use any opportunity to go mountain climbing with my friends.
I have conquered the Damavand [elevation 5,609.2 m (18,403 ft) ]
and Sabalan [elevation 4,811 m (15,784 ft)] mountains.
Sapienza Università di Roma
Rome, Italy
E-mail :
Please replace ** with "R" span> ;)
Tel:
Skype ID : hashemi_hossein
Cell : +39 (351) 928 0194