RIT 2017 Internship Experience

Today has been a very slow day. The majority of the day before Kamran arrived i had little to do as my structure draft had not been sent back to me with a review by either Kamran or Rakshit. In the time i had i attempted to locate and edit video evidence to support my statements in my presentation. I had luck finding the videos but editing them was another problem as i could not find any free editing software that could actually crop and cut a video. Unfortunately i had to call in a favour from Aditi as her Mac comes with built in editing software. She has edited some videos for me but i hope i can get Kamran to teach me a better method so that i don't need to bother Aditi's work. Kamran finally arrived and we went over my draft, with him giving me some advice on how to refine it. The rest of the day was allotted to my draft.

Modelling Human Hand-Eye Coordination Using Recurrent Neural Networks [TM1]   I        Introduction What the presentation is going to discuss/ introduce the project/ keywords [TM2]   II     Rationale o     The goal of this project is to create a VR game in which participants catch a ball, and there head, eye, and hand position/orientation are recorded. Using the participant’s data, a reinforcement learning algorithm will be developed to create a self-learning AI [TM3]   . The AI’s performance will be compared to the performance of the participants via simulation in a virtual space. o     The goal of this project is to create a VR game in which participants catch a ball, and there head, eye, and hand position/orientation are used to develop a self-learning AI, which performance will then be compared to a humans in the task of intercepting a ball. III     Backg...

Today was a very productive day. In the morning until lunch i worked solely on refining my presentation into a draft i could send to Kamran and Rakshit. After lunch the rest of the day was allotted to another session of mathematics with Rakshit in which i learned more about signal based mathematics and actually had a mentor competent enough to explain it to me in a way that it didn't take me an entire term to understand. After that Rakshit gave Aditi and me another mini-assignment of distorting the pitch of a sound file. I've found a royalty free sound file and plan to edit it tomorrow, depending how it goes i'll use it someway in my project.

Lo and behold another windows update that will completely destroy my ability to blog as the update took 3 hours and at that point my CPU was hot enough to cook an egg and it was already an hour past my bedtime. As i recall, the events of yesterday were mainly based around spending until noon analyzing the last remnants of eye tracking data, detecting any common tends or patterns of motion, which we will then show at the weekly meeting on Thursday with evidence to back up our claims. After lunch i "borrowed" a lamp, curtain, and PCP piping to attempt to recreate the effect of the virtual experiment as i created a crude cover to block my participants line of site temporarily as i tossed a ball. If the data is sufficient i will attempt to compare the eye movements of both virtual and real tracking of a ball, and present my findings in my final presentation, as well as show the inefficiencies of the real experiment, enforcing my claim that physics based experiments performed usin...

The day has been very quiet as Aditi left the lab early at 12:30 and no one has been in the lab besides me today. I've primarily been working on my structure, but also did some work defining eye patterns with Aditi before she left. At 10:00 i left to go to the symposium, i could not find the specific poster i was looking for no matter where i looked but i did look at many others. The VR integration of a Mac was particularly interesting and held marketable concepts. I was only able to talk to one of the presenters as the others did not speak English very well, although i was able to understand some of the words they said from the very little Chinese i know. They began whispering to each other when i approached and i could hear the word "talk" which explains why soon after the presenter approached me. The rest of the day will be dedicated to improving my structure and possibly collecting some data from Kamran to support a comparison for my final presentation, but i will lea...

The entire day was comprised of data labelling for Rakshit. I allotted the entire day to labelling so that i can allot all of tomorrow to work with Kamran and hammer out a lot of work before the weekend. Their was one moment of reprieve today during the floor meeting where Jeremy did a mock presentation for his symposium tomorrow. This gave Kamran the idea that Aditi and i could also do a mock presentation in front of both a semi-lay audience comprising of brilliant minds who can assure our presentations be perfect for the final day.

Today was a slow day to start out as i repeatedly attempted to draw a vector from the cycEyeNode to the ball in the simulation using viz.primitve. Unironically i finally figured out how to do it on the bus home. The later stages of the day were all spent towards collecting more data on Henry and Nate, both of which provided very interesting data from a broad look at it, as by the time i could zip the files, transport them via Gdrive, and unzip them, it was 20 minutes to leaving time. Overall a good day, and i'm starting to feel my internal alarm clock going off telling me that the internship is ending soon and i need to tie up all loose strings.