Mtg 3/25: Mon-11-Sep-2023

Outline for Today

Wiki

Link to the UR Courses wiki page for this meeting

Media

Transcript

Audio Transcript

  • Okay, so recording
  • anyone not have one of those or want to make one make a new one.
  • So fold it in half and then fold it again
  • please draped in large letters not in not using a yellow highlighter your name the name you'd like to be called so not your full name, just the name
  • that I should work on remembering Okay
  • oh I shouldn't read on this board because I'm gonna pull money
  • okay so wish you Happy Monday
  • maybe it can't be more than one thing at once. So let the display Come on then I'll switch to HDMI
  • Okay, so how many people have been on your courses
  • now, not right now but it would help it wouldn't. If you haven't answered the question about when to schedule the midterm go to your courses now. And maybe I have to
  • increase
  • damage
  • okay
  • okay so three people haven't answered yet I'm gonna extend it a little bit more
  • yes
  • okay you have 10 more minutes if you haven't done it, do it now
  • so can everyone find this
  • I would like to have a brief discussion about your courses because with the last meeting response there were only six six out of 13
  • So this is just an easy way to keep engaged with the class and it's an easy way to lose 5% of your final grade
  • if you don't do it so I want to make sure that it's clear what's happening and where it's located and so on and so forth.
  • How many people do we have here? So we have 10. So it could be that the 10 of you? Or the 10 who've answered any no shame in this or Has anyone here not answered the poll?
  • Know 12 or 13? Either of you.
  • So this is my first day first class. I don't know how it goes but today.
  • Okay, well take a minute to do it. Now. You've been in your courses before, haven't you? I said your absolute first experience. Yes my
  • last I learned this last night.
  • Well welcome. So do you have you sent me email? So you can log in, gotten your account set up so go to your courses duck theory chinese.ca?
  • Sitting there for the first time
  • so your email that should have indicated to your username being set up
  • so what if you go to you you are courses dot Uri child.ca and type in credentials you set up your email
  • this page you have your route with us. Yes sir. With the same user ID and password for your business your number then you can change it.
  • Meet
  • these groups.
  • On this is
  • it on okay we'll work on that later Okay
  • Oh Mark your attendance either today. Everybody else can you access that on your courses
  • so
  • if you go to your courses I just want to make sure that I'm not what's the word making this more difficult or glossing over some things. So does your courses website and then there's my website that links to the URL courses stuff
  • so this is the new art courses website.
  • So when you log in
  • this is what you'll see when you are courses. So go your courses and you'll be set and then there's a link to my website
  • so then under meetings so there's a list of meetings. And then in general there's a link to my website.
  • So if you go to the List of Meetings page you can click on today's meeting. And that'll go to
  • my web page with the outline for today. Let's see. Okay. So read the other quick link I want to show you or remind you of
  • is happening teaching today and that'll have the link for today's meeting. So you can go there first, but to access the attendance and stuff you'll have to log into your courses
  • I'm not going to make a big deal about this in future I'm going to consider that it's been covered now. If you don't want to take your mark your attendance or write responses that's your decision. Okay. I just want to make sure there wasn't some communication issue
  • I wrote this last night so those 1020 So we'll look at it now.
  • Well I guess Monday the 23rd Has it
  • so one more vote anyway
  • won the popular vote Okay, so Oh, just just make note of the midterm in in the outline and stuff
  • So I want to make sure that I write down not quite on the board. So let's talk a little bit more about the pixel.
  • It's not a cross product or multiplication sign. It's just an ethics
  • okay. So it's a physical. So you can think of as a container, maybe that stores value but it doesn't have to be physical. I mean, we could think about
  • it then it's going to some part of the integrated circuit that's in the computer in the computer's memory
  • made maybe it's better as a conceptual a logical container
  • so I found a couple references. This is about a four or five page document written in 1995 or 96 by ldra Smith, who co founded Pixar. This is when he was at Microsoft pixels not a little square pixels, not a little square pixels, not a little square and the box was not a little cube
  • you just
  • saying same construction.
  • And then I realize he's written a book called a biography of the pixel, which is sort of a history of computer graphics published by MIT Press so I gave you the link to that one as well. That would be interesting to read.
  • So the idea is
  • a pixel is a sample
  • now we just started talking about.
  • So we can think about those two things. In terms of images that we want to create with a computer so with the starting point
  • what are the
  • How can we describe an image in terms of samples or sampling and quantization?
  • So if we think of a picture we take on a smartphone how do we describe it or what information do we have about it?
  • Date and diamond was taken.
  • So we have meta data
  • how much every was size size
  • in settings
  • so is that
  • maybe also part of the metadata
  • okay, I prefer image like to BG
  • image format?
  • Joined picture experts group that that's a trivia question. When you're out with friends, do you know what JPEG stands for?
  • So it's a lossy compression format. So it's lossy
  • that's
  • yeah. Anyway, that's my joke. So what happens is when we do a JPEG image compression then we're taking out some of the higher frequency samples so we're we can encode it more compactly so when you're saving a JPEG format, you can say high quality or low quality you can have a
  • you can have a say and the amount of compression so you can make a very highly compressed image that's clearly not the original image or it can make a somewhat compressed image which was a fairly good representation of the image okay.
  • This same thing what pixels when you look at an image so sure
  • so that's the spatial resolution of the image
  • Okay, so weather try
  • whatever the choice we have about images neither know where this is reflected in in the information that's stored presented
  • Good question. Yeah. I was just looking at a picture taken. It has 12 megapixels. Are we taught is that also going to be spatial resolution or is that the number of pixels in the image in sets? So 12 million. Pixels
  • Yeah.
  • Now it's a little bit so it's a little imprecise. Thank you for giving me a chair
  • so a megapixel could be a million. So when we're talking in decimal we use shorthand of a mega for a million. But if we're thinking about binary it's 1024. squared, which is not quite a million. So to be clear about that, if we want to be precise, but it's the width times the height
  • that gives you the number total number of pixels so 12 megapixel sounds better than 3820 by 24. Or whatever. That's a better shorthand maybe.
  • So we have the size of the storage the image takes, and that'll be affected by the image format. We could use if we did PNG or if we store it there are some savings in a PNG format even though it's lossless. If we just had a raw format, the data would be the number of pixels times the pixel bits per pixel. That would give us the amount of storage for the image.
  • So when we're dealing with this, it's not we're not in theory, that's how we relate the numbers but
  • the math won't work out. Exactly, especially if we have JPEG images. That that's the idea that we're storing so what are the bits per pixel represent?
  • I don't know if I have a feeling I took this already been
  • edited in post production
  • So we said pixels a sample. So what is the blank of each sample? And we're talking about the number of bits that we're storing with each pixel.
  • That question makes sense.
  • Could you repeat that please?
  • Okay. So pixels are samples. And we're talking about the number of bits per pixel that we're allocating. So we're restoring something about the sample that is the corresponds to that picture element.
  • So how many values can we store for a sample if we have one bit per pixel?
  • What what binary numbers can we represent with one bit one binary digit we can do is zero or one.
  • So it could be a very high spatial can have very high spatial resolution 12 megapixels.
  • But if we're storing one bit per pixel it's gonna be black and white. So the pixel values are going to be black or white.
  • So it doesn't give us a lot of detail I'm gonna say this in a very badly so please forgive me. There's a story of someone who did a great service for a king and the king said, What would you like for your reward? And he said, I would like one grade of race on the first square this chess board and then double that on the rest of them. A king thought that was a good idea. But by the end I ended up to a lot of race.
  • Laughter where if you can find a nice version of that story, online, but the idea is when we were using binary like we're doubling each time somehow that's
  • it. So 24 bits per pixel we get a lot of expressive power in terms of what we can represent
  • So, does that make sense? So, we have samples the number of samples is a special resolution and then the number of bits we're storing per sample is.
  • is, I guess the resolution of values within associated with that sample?
  • Okay
  • Does that make sense?
  • So it's not only in computer technology that we have discrete the idea of discrete representations. For example, when we see a film.
  • So when there actually were projectors of filming theaters, now they're digital projectors. They give you 24 They will be 24 frames per second. So we're taking 24 still images and we're showing them the high enough rate that we get the impression of movement.
  • Think I was doing all of my writing I'm getting worse
  • Have you seen mosaics of tile where the tiles are the same size but they're put together in such a way that you get the impression of an image
  • so I'd say
  • That make sense. So we want to sample it twice. The highest frequency that we want to reconstruct?
  • Okay, so we looked at some images, we zoomed in.
  • Save real life. is continuous we make discrete approximations.
  • Do one more blackboard here
  • Evening Sorry I just getting two readings being too slow there. So once we have a discrete bitmap image
  • and it says it's time for me to stop writing on the board. So once you have discrete bitmap images, zooming in will show pixel artifacts. So even though we have something that looks very nice
  • if we zoom in
  • that's not supposed to be exactly but gives an impression
  • we can see that.
  • So we have been out funds for example, we need to store them in different resolutions so that the look at the look at the different sizes we want in which we want to use them.
  • So, vector images people are you familiar with
  • was gonna say CSV images but that's not right is SVG. Scalable Vector Graphics supposed to comma separated values?
  • Storing a mathematical description of the image
  • and so instead of
  • so we can zoom in with SVG images, for example. Because we're not storing the Bitmaps we're not starting to edge individual pixels. We store the description and then we recreate the the we sample the mathematical description at the requested resolution. So if we were to save that image at a particular resolution, then we lose the benefit of the SVG format. In terms of being able to rescale it
  • because that's that's a way that we can get around the limitations of storing about bitmapped images.
  • mean if we had if we have I mean they're not always limitations but because if we spend two days rendering an image at a high resolution firm moody frame we don't need to store that we don't need to store that. That we don't need to regenerate a different image at different resolutions so so bitmap images aren't bad. You just have to understand what's what's going on with them.
  • Are five minutes left here I wanted to just show you that. I didn't get the link made.
  • This is a syllabus that I
  • use are linked to our courses page. Office hours Tuesdays and Thursdays 10 to 1130.
  • So basically, our outcomes for the class are to understand basic concepts of computer graphics and animation. So demonstrate your ability to play these techniques in software that you write and understand how to apply the graphic animation frameworks in your work. So it will be very with some hands on with a good dose of hands on.
  • Here's the information on the text. Here's the link to the website associated with the text
  • and we talked about this already assignments 40% participation 10% midterm exam now on October 23
  • And the final exam 40% That's on December 13 I believe.
  • December 13 at 9am So both the midterm and final will be written exams in person. likely they'll both be in here. If we don't if you don't like this for the midterm then we can find a different place for if the comfy chairs are too comfy for exam writing. Then we can find a different spot and here's kind of a breakdown. We're going to go chapter by chapter
  • so our first assignment due at the beginning of October and we'll talk about assignments next week. Get your first assignment sorted out next week. So it's basically two weeks per assignment. As a bigger space here because of the midterm in here
  • so if you're late and be penalized percentage of your assigned grade. If you missed the midterm, that's that way to be put put onto the final exam. So be 50% final instead of 40%.
  • So it's in your interest to make it to the midterm exam. If you missed the final you get an NP so attendance is expected you can record your own attendance in your courses. There's about academic integrity, some standard stuff.
  • accommodation. So I'll post that right after we're done here. Let's see. Okay.
  • Doctor the link for the syllabus on the on your website. It's broken. So it doesn't yes
  • I'm gonna fix it right? Momentarily so
  • what else do I need to say? So today when I have more luck with three Gs for who's got it figured out? Working for that simple example. Okay, so we'll, we'll do some programming next day. What else one minute left.
  • This is one of the things I did all weekend was to add links to that slide
  • Okay, so I just got these two lines together so it didn't show up. I'll regenerate the file so you get the link for the response will show up and they'll show up say, find his name before 11pm tonight. And so just a reminder that I got only six last time so I'm going to do that right now. Before I leave here, okay. Anyway, thanks for today. See you on Wednesday. If you have questions or concerns come see me when office our
  • office hour and a half and we'll see you Wednesday Have a good rest of your day. Thanks bye

Responses

What important concept or perspective did you encounter today?

  • It was really nice to know the full form of JPEG and what pixels actually mean
  • is about the discrete bitmap image and how the jpeg image compression works
  • From the discussion, I grasped more clearly about the difference between bitmap and vector images: Bitmap images have fixed resolutions, making them susceptible to quality degradation upon zooming, whereas Vector, described mathematically, retain clarity regardless of scaling.

Was there anything today that was difficult to understand?

  • In today's class, it was hard for me to understand continuous vs discrete. I did not get the significant difference between these two. It would be great if you could give more example on this. It was also hard for me to understand the perspective of continuous & discrete in the field of computer graphics. Also if you could tell us before hand or after the class that which chapter from the book are your lecture from, so we could do some reading in order to relate to the lecture. Thank you.
  • I understand what professor wrote on the blackboard up to the explanation about mosaic. The rest was a bit difficult for me especially on the parts of svg(vector) and bitmapped.

Was there anything today about which you would like to know more?

  • I learned that pixels aren't squares but abstract concepts in digital imaging, and the meaning and working of continuous vs. discrete representations. Apart from that I also amazed by the mega pixel by bits conversion which is very interesting to me.