PhosInv: Developing an Optimization Model for Visual Cortical Prostheses Using Deep Learning
It has already been a year and a half since I began my undergraduate research at Professor Seung Woo Lee’s lab at KAIST early last year. Starting from scratch and studying hard at whatever I could, it seems that some results are finally coming out. In fact, since I am still revising the draft, publication is still a long way off, but at the same time, I harbor the hope that we are already reaching the final stages. I sincerely hope that it will be successfully completed as a full-fledged paper.
Professor Seung Woo Lee’s lab in the Department of Brain and Cognitive Sciences has the goal of developing visual cortex prosthetics. To ask what visual cortex prosthetics are, have you heard of Neuralink? It is one of the technologies used to restore vision to the visually impaired; it involves inserting electrodes into the visual cortex located in the occipital lobe—that is, directly into the brain!—rather than the retina, to apply electromagnetic stimulation and enable the perception of visual patterns. The realization of “deep links” and “full dives” seen in virtual media is not far off. (This is a joke)
Anyway, I wanted to complete something within my undergraduate studies in this lab, so I applied deep learning and did some coding. After investing a year, I did produce results. To put it simply, the content is that while previous research optimized variables for inserting visual cortex prosthetics, my methodology is more practical and efficient than theirs. Since the purpose of the technology has changed, it is difficult to make a direct comparison, but I tried to weave it together well to create a story.
Since my work even after coming to New Zealand has been fixing figures and revising papers, I felt a bit unsatisfied, so I decided to turn this into a video. Based on materials from a lab meeting held just before I came here, I tried to recreate the presentation by ruminating on my memory of it. Of course, I spoke in Korean back then, but this time I made it in English. There are Korean subtitles, so I recommend you give it a watch without worrying. Unfortunately, I cannot take responsibility if it is not entertaining.
I will become a graduate student starting next semester, and I am worried that I might not be able to write or upload videos occasionally anymore. I hope I don’t lose even the few hobbies I have left. For now, I should finish my current paper and try submitting it to the conference. The deadline for this conference is in September, but they still haven’t posted the official announcement. I should at least get ready in advance.