User Profile
Add Friend
Add Note
Track User
Send V-Gift
Matt in Cambridge
Halcyon Days Across the Pond
Created on 2004-09-27 00:29:22 (#4660637), last updated 2004-10-18
1 comment received, 0 comments posted
Basic Account [Gift]
3 Journal Entries, 0 Tags, 0 Memories, 0 Virtual Gifts, 1 Userpic
| Name: | Matthew Johnson |
|---|---|
| Birthdate: | 01-05 |
| Location: | Cambridge, United Kingdom |
| Website: | http://www.cs.arizona.edu/~mjohnson/ |
I'm 23, from Toronto, Canada by way of Phoenix, Arizona, where I lived for 14 years. My life has always contained a dichotomy of art and science and as a result my choices have always been affected by these two forces within my soul. It has and will always be my goal in life to pursue a career which combines these two facets to create a better world for myself and those around me. As for this dichotomy, it stems from my parents, who are two rather different people. My mother, Eleanor was a musical prodigy and is a gifted singer, pianist, and organist. My father, Graham, was and is a gifted mathematician and computer scientist. They have four children including me, which include Colin, my gifted organist/accountant brother, Elizabeth my politically-minded sister whom I adore, and my underappreciatedly great youngest brother Peter. I am the eldest, with all of us two years apart. All of us, from very early on, have lived a dual existence of left and right brain, science and art and I, for one, have always tried to maintain a balance between the two. However, after graduating from high school I was faced with a choice: whether to pursue a career in music or in science. My mother advised me that while music was a wonderful thing, it made a better hobby than a career. Based on that advice I decided to enter into a scientific discipline at the University of Arizona.
I signed up as a mechanical engineering major, but within two semesters I had switched from mechanical engineering to computer engineering to computer science. I delved into computer science with fervor. It contained a mixture of linguistics, science, and human interaction that was a perfect mix for me, and I loved every minute of it. However, as time passed, I found that I loved all of my new field equally, which presented another problem, namely, what I would do for a living. Now, I was relatively certain that I wanted to remain in academia. I’d been teaching first as an undergraduate section leader and then as a TA for 3 years and knew that I loved to teach. In order to become a professor, though, I needed an area of research, an area which was of true interest to me. I couldn’t find one, and as such I floundered for a year or two, when I met Kobus Barnard and was introduced to Computer Vision.
Computer vision, in short, is the quest to enable computers to see. More specifically, it is the search for methods of garnering information from images. Now, this may not seem like that big of a problem, but that is because human beings are incredibly good at seeing things. So good, in fact, that we do it unconsciously and automatically whenever we open our eyes. A human being can look at a room and know immediately what objects are in the room, what people are in the room, which of those people they know, what those people are eating, whether those people are male or female, and so forth. Not only can they do this, but they do it almost instantly. It is truly astounding when you consider how complex and difficult vision really is. For example, take the task of segmentation. When you segment an image, you break it down into constituent parts, or segments. This alone can be incredibly difficult. You might think that you can do it by pattern, perhaps by color. However, think for a minute of a penguin sitting on an ice floe. You have a large area of white (no patterns), a smaller area of white (again, no patterns) and then an area of black (yet again, not a pattern to be seen.) Now, somehow the human brain can figure out that the white and black patches are in fact parts of the same animal that is also distinct from the surrounding ice. Not only can it do that, but it can name the bird, make predictions as to what direction its going and how cold it is. The human brain is a frustrating benchmark for researchers in computer vision.
After completing my bachelor's degree at the University of Arizona, I matriculated to the University of Cambridge, England, where I will obtain my PhD in Engineering Sciences with a focus on Visual Recognition. I began my blog with the intention of keeping in touch with everyone back home in Phoenix, Arizona. To that end, I will document my travels and travails through the exciting corridors of academe. Dullarity ensues.
I signed up as a mechanical engineering major, but within two semesters I had switched from mechanical engineering to computer engineering to computer science. I delved into computer science with fervor. It contained a mixture of linguistics, science, and human interaction that was a perfect mix for me, and I loved every minute of it. However, as time passed, I found that I loved all of my new field equally, which presented another problem, namely, what I would do for a living. Now, I was relatively certain that I wanted to remain in academia. I’d been teaching first as an undergraduate section leader and then as a TA for 3 years and knew that I loved to teach. In order to become a professor, though, I needed an area of research, an area which was of true interest to me. I couldn’t find one, and as such I floundered for a year or two, when I met Kobus Barnard and was introduced to Computer Vision.
Computer vision, in short, is the quest to enable computers to see. More specifically, it is the search for methods of garnering information from images. Now, this may not seem like that big of a problem, but that is because human beings are incredibly good at seeing things. So good, in fact, that we do it unconsciously and automatically whenever we open our eyes. A human being can look at a room and know immediately what objects are in the room, what people are in the room, which of those people they know, what those people are eating, whether those people are male or female, and so forth. Not only can they do this, but they do it almost instantly. It is truly astounding when you consider how complex and difficult vision really is. For example, take the task of segmentation. When you segment an image, you break it down into constituent parts, or segments. This alone can be incredibly difficult. You might think that you can do it by pattern, perhaps by color. However, think for a minute of a penguin sitting on an ice floe. You have a large area of white (no patterns), a smaller area of white (again, no patterns) and then an area of black (yet again, not a pattern to be seen.) Now, somehow the human brain can figure out that the white and black patches are in fact parts of the same animal that is also distinct from the surrounding ice. Not only can it do that, but it can name the bird, make predictions as to what direction its going and how cold it is. The human brain is a frustrating benchmark for researchers in computer vision.
After completing my bachelor's degree at the University of Arizona, I matriculated to the University of Cambridge, England, where I will obtain my PhD in Engineering Sciences with a focus on Visual Recognition. I began my blog with the intention of keeping in touch with everyone back home in Phoenix, Arizona. To that end, I will document my travels and travails through the exciting corridors of academe. Dullarity ensues.
Interests (15):
Friends [View Entries]
Communities [View Entries]
Feeds [View Entries]