News and Updates
Five New Papers Published in Recent Months!
In the past few months, I published five new papers. They are related to active robotic vision, human-robot interaction, and human eye's retinal structure prediction via machine learning. You may want to check them out!
An Internal Transition: I Joined Amazon Robotics!
In the end of June 2022, I internally transferred to Amazon Robotics to move to Boston area and work on a wealth of exciting computer vision problems to be solved! Focus of my work in the new team is on developing machine learning and computer vision methods to automate tasks in Amazon warehouses. I also truly appreciate my former team members for having the opportunity to learn from them and building great products with them.
I joined Amazon as an Applied Scientist
In February 2022 I joined Amazon Lab126 as an Applied Scientist. My work will focus on developing sensory algorithms, based on machine learning and signal processing, for different products being built in Amazon Lab126. My move to Amazon coincides with the conclusion of my postdoctoral research work at UC San Diego. The time working with Drs. Welsbie, Zangwill, and Christopher in the UCSD's Department of Ophthalmology was an invaluable experience that helped me gain first-hand insight about different aspects of a biomedical imaging project. I would always be grateful for the friendly research environment that they provided. Now, I look forward to start a new chapter in my career at Amazon!
I passed the Tensorflow Developer Certificate exam
On September 12, 2021 I took the Tensorflow Developer Certificate exam by Google and passed it. It was a 5 hour long test with practical coding questions to develop and train classification and regression deep neural network models for computer vision, natural language processing, and time series data applications.
Our article on next best view has been published in Springer Applied Intelligence
My latest paper has been published in Applied Intelligence journal. It is one of the papers originating from my PhD dissertation. It presents a next best view method for active robotic vision, specifically for active object recognition. To test the method, a dataset suitable for evaluating next best view systems was gathered by me, with the help of my PhD advisor, Dr. Mircea Nicolescu. You may check it out at my GitHub page.
I joined University of California, San Diego as a postdoctoral researcher
Recently, I started a postdoctoral position in the Department of Ophthalmology at University of California, San Diego (UCSD). In my research, I will work on using meta-learning and semi-supervised learning methods to predict the thickness of ganglion cell inner plexiform layer (GCIPL) and retinal nerve fiber layer (RNFL) of macula. These predictions can be useful in treatment of glaucoma through neuroprotective methods, and especially in reducing the cost of larger-scale clinical research for this treatment approach.
I believe, working as a researcher at UCSD would a rewarding and constructive experience!
Honored to receive the Outstanding Graduating Graduate Student award at UNR
On December 7, 2020 I was selected as one of the three awardees of the Outstanding Graduating Graduate Student award at University of Nevada, Reno. This award is given annually to three last year graduate students who have a distinguished record within their program, while having shown leadership qualities on campus and have worked for the improvement of the university.
PhD defense successful; Doctorate conferred
On August 28, 2020, I successfully defended my PhD dissertation, entitled "A Single-Shot Next Best View Approach Accompanied by a Dual-View Active Vision for Object Recognition Tasks." As the name suggests, I developed an active vision framework along with a viewpoint planning mechanism that enable robotic systems to effectively improve object recognition performance.
It took 5 years of hard work toward performing research, teaching, studying courses, and authoring publications to reach this pivotal moment of earning the PhD degree of Computer Science and Engineering from University of Nevada, Reno. Now, I am officially a doctor!
I was chosen as the Outstanding International Graduate Student at University of Nevada, Reno
On May 8, 2020, I had the honor to be selected as one of the Outstanding International Graduate students of University of Nevada, Reno in the academic year of 2019-2020. This very competitive annual award is accompanied by a scholarship of $1000. The few recipients of this award are selected based on their scholarly work, scientific leadership, and extracurricular activities.
Our paper features the cover page of the Computers journal
Recently, the paper we published in Computers journal was chosen by the journal editors to feature the cover page of the quarterly in December 2019.
The paper proposed a novel active vision system for humanoid robots to increase the performance of object detection by dynamically utilizing two cameras to get different viewpoints of objects. One camera is on the robot's head and the other is installed in the forearm of the robot. Whenever there is an uncertain detection in the viewpoint of the head camera, the arm of the robot and the camera on it move toward the object to get a new viewpoint of it. The camera system automatically adjusts its pose towards the objects with uncertain detection. Later, after a matching stage, the detections from the two camera views are fused together to form the final results.
How I got interested in AI...
I want to start off this page with my personal story of how I chose to work on doing research on intelligent systems.
My journey in the field of machine learning and generally in artificial intelligence started in 2008 when the instructor of one of my graduate-level classes, and later my advisor, Dr. Khoei, distributed a three-page handout. One page about neural networks as a representative of machine learning methods, the other for genetic algorithm on behalf of evolutionary algorithms, and the last one about fuzzy logic as an expert system. After reading those brief notes, it was the point my fascination towards AI began. How enchanting is to have a system that learns, or evolves, or adopts the existing knowledge?
It was a turning point. Later, I learned other machine learning and evolutionary algorithms and grasped deeper understanding of how they work. Now, I am proud that I am a PhD candidate in my beloved field of research.
My online nest up and running!
pouryahoseini.com is officially live! š„³
