Joshua Situka July 29, 2009
This is probably my last blog, unless I feel compelled to share some more of my feelings. I would probably recommend this program to other people on the simple fact that they do alot of self-evaluation and are always trying to improve themselves. With all the things they have to worry about, and from what I hear about other REU's that is pretty impressive.
Today, we should act like we are doing work but really there is nothing to do. Linh sent me his stuff which I will combine with mine then send to Song. Thats it. We are ready for the demo, and to leave. Ames, Ia aint that bad, i think you have to search harder than most places to find something to do. I learned long ago that its not the place that limits your entertainment its your imagination that does. So even if this place isnt in an ideal location, I would consider coming here for grad school. Only thing is my imagination cant change the weather.
I think I got Kelly hooked on Typing Maniac, i have a feeling she is trying to beat my score.
I wonder if I'll keep in touch with my teammates after this. Prolly not phone calls. I assume they like me but I don't actually know enough to keep in touch. Also, learned long that if you don't know assume the positive. But if I ever beat Linh's Tetris score, or if Kelly ever beats my Typing Maniac score than I have a feeling that there will be some words between us. Maybe Zayira is right, Facebook might never go away, perfect way to casually keep in touch.
Joshua Situka July 28, 2009
Still revising the paper nothing special today. Yesterday Song showed us his $30,000 camera that had a ridiculous megapixel count. He also took 3D video sequences of us with his camera. He seems very happy with what we have done, which is 180 degree turn from how everything stood just a month ago.
Also, best and worst things this summer. Okay worst was definitely STAT, the teacher was cool but the timing and subject matter was just a good recipe for me to fall asleep..on numerous occasions. If I would of have a tutorial of
OpenCV that would have been great but I dont think many other groups would have use for that. Maybe a class on getting money for graduate school would be more appropriate, so that we would have more tools available.
Best was journal club. Zayira let us be pretty open with our topics and most were HCI related. I feel that discussions are always fun and I like hearing other's opinions especially ones that differ from me because it always gives me something or another point of view to think about. I know everyone else is not like that but i like conversation..in case anyone didnt notice.
Joshua Situka July27 2009
Today, I just revised the paper again and again till Song's happy. Such is my life.
And we are just getting ready to do the live demos.
Joshua Situka July 22 2009
I have been counting down they days to my b-day, and i just realized...im just going to be a day older tomorrow than I am today. 22 is not quite as special as 21. oh well, good excuse to celebrate anyways.
Project is coming along okay. Just doing the paper. I am going to add in pictures to the results and methods. And that's it ..for the first draft. I already know its going to get butchered, and endlessly critiqued, so im not stressing. I revised my poster 10 times at least so im guessing the paper is going to get at least 20 revisions before my mentors are cool with it.
Transformers tomorrow, on IMAX, yay!
MinaChoi - 25 Jul 2009 - 22:25
Happy BIRTHDAY SPAM!
Happy Birthday!
Joshua Situka July 17th 2009
So went to Ken's house yesterday. Needless to say it was an interesting experience. The guy has 3 parokeets, a dog, and two turtles. And He loves to play with all them. It was funny watcing him play fight with his dog (Emmie) because he would cut the legs out from under her and she would land upside down, and then he would bear hug her. And if that wasnt enough Emmie would be picking fights with the parokeets who are 100 times smaller, and not win. And then Linh is trying to work on the Zcam program but the parokeets keep knocking it down, and the proceed to climb over his laptop (as Ken would say "they like to type").
Then me and kelly were going at it with Super Knock out (a cool boxing Wii remake of a classic game), and she got pretty much into it, but I still prevailed 10 wins to 6 wins. She says that I was cheating but throwing parokeets at someone's back, covering their eyes when trying to play, and taking away the controller..doesnt really qualify as cheating in my book. I played completely fair. I just play to win.
Oh yeah, we are suppose to talk about work too. Im going to finish the poster today. Again. with Song's suggestions.
How long is your results section?
mmmm.... it sounds to me like cheating....
Joshua Situka July 15th 2009
8 days till 22. Yay. I think i might of got the calibration code working like I want to. I wanted to zoom in the window, convert from BGR to HSV, and use three colors to track the skin detection instead of two. They are all completed.
Also, showed Song my poster yesterday, he pretty much hated it, therefore i did a new one today with all the bells and whistles. I really dont want to show him, because I dont feel like making a 4th draft. Kinda want to surprise him on Demo day.
Also, if anyone is looking to take out some aggression Fight Round 4 is the perfect way. You get to hear the bones crack, and see the face flap in disgust when you land a haymaker. So I was facing off with Ugonna. He was Frazier and I was Mike Tyson. However, I was imagining his fighter's head as the Zcam. Needless to say, i won every time.
Never worry. One tutorial and GAME ON BRO!!!
Joshua Situka July 13th 2009
10 days till 22. Yeah.
Anyways, we went paintballing yesterday. And i was slippin and sliding into bunkers and trenches giving everything I got. I ended up screwing up my ankle for the rest of sunday, but it was worth it. So I had fun.
Back to work today. It's one thing to get something working,...to perfect it. Perfecting the Zcam calibration is annoying because now it's "working" but doesnt get me better values than what it was before i tried to perfect it (Which defeats the whole purpose). So yeah, that and HSV, and putting in three colors for comparsion as suppose to two are what I am working on today.
Yeah, I'm still really sore from all the sliding and diving on Sunday, but it was worth it. Hope your ankle feels better soon.
Joshua Situka July 10th 2009
So this guy Lucas is doing his master dissertation over 3D modeling of the hand using the Zcam at 1:30pm. I think im going to have to miss journal club because this sounds way too promising for me to miss out on. Linh seems to have a better tracking of the fingers using some vector thing Ken told him about. (everything that Ken touches seems to just work magically). Kelly is still slowly improving the finger movement. Song thinks that the calibration of the camera will fix it, but im not too convince. For some reason the camera will track the hand but it needs a longer IQR range. which is kinda hard to explain but basically i have to change the code but then it will only capture skin color, and one more color.
Results are due today, Hah....everyone else just started their "experiments". We'll see how far we get. I'll maybe put something down for how far we got right now.
Joshua Situka July 7th 2009
I managed to calibrate the camera, now I just want to make it better. Brett showed me some way to use the opencv function cvSetImageROI to put the window a certain way. the program crashed. Then I tried my own way. The program crashed again. Programming is very fun sometimes. So me, Kelly, and Linh were thinking of re-enacting the the office space scene where they beat the crap out of the printer with a baseball bat. Instead we have the Zcam. Unfortunately the zcam cost $600 bucks, and we prolly cant afford to replace the its box. So i think i might just take a picture of the Zcam. And give Kelly the bat and let her go to town, she doesnt say it but I can tell she would like to go a couple of rounds with it.
please look at the email Pam sent and sign up for our study
MinaChoi - 08 Jul 2009 - 23:45
lol hamadi.... JOSHUA COME BACK!! D=
Well you know what they say. If at first you don't succeed. Try, try ... try, try, try ... again.
yeah, those ZCams are fun to play with at first, but after working with them for a little while you realize just how ridiculous they are.
its calibrated! that's great! please don't pull an 'office space' hehe
Joshua Situka July 6th 2009
They said they want to hear about our feelings on this blog okay here they are. Right now i'm frustrated that we didnt get the checks direct deposited. now im jumping through hoops trying to pay America Airlines $400 for their bereavment fair to go to my cousin's memorial service. My bank apparently needs 3 days to authorize the check...but guess what in three days is when the memorial service is. My parents dont have the money because they just found out they are the "next of kin" listed under my cousin and since my uncle was in a different continent they are already taking care of flying the body back to uganda. (thats $3000)
I may be an extrovert and have lots a good acquintances, but really there is only 5 friends I can call and count on them helping me pay it. I called each one and the last one, fortunately, Steph said she will help. But still i'm not sure if I qualify for a bereavement fair to begin with because every other airline apparently "cousins" dont count. Im hoping America Airlines is somewhat compassionate considering im the "next of kin's" son...whatever that means.
So right now im frustrated, and just tryin to get this zcam to calibrate correctly in which in refused to do during the 6 hours i was here yesterday.
PamShill - 07 Jul 2009 - 07:25
Sorry, Josh. Please let me know how it all turns out. I suggest that you contact your grad student mentor to help you with the zcam calibration. Sounds like you have spent enough time on this task and could use some help or suggestions.
MinaChoi - 06 Jul 2009 - 15:45
=( wow.... I'm sorry for your losses... I hope things work out...
I'm sorry about your cousin Josh, and that sounds very frustrating
Joshua Situka July 1st 2009
I am trying to do this image processing with openCV, and yestereday took me 3 hours just to get the picture to be recognized. You have to include all these dependicies, library, and include files in 5 different parts of Visual Stupido. I wanted to drop kick my monitor when the same errors popped up, everytime I THOUGHT i fixed something.
Today things running a little more smoothly because I got the color count of the pixels, now I just have to compare to Matlab to make sure the codes produce the same thing.
Then off to calibrating colors....which i really have no clue how to start.
Countdown 22 days till im 22. Yeah the most unexciting b-day ever....There's really no more to look forward to, except when i turn 55 and get senior discounts at IHOP. That will be a Great birthday,only 33 years and 22 days to go.
Joshua Situka June 29th 2009
The Zcam is frustrating me. It works perfectly one day, and today i spent two hours just trying to make the colors work again. Now they do..but I don't even know what I did to fix it. The colors are very sensitive.
Song has decreased the scope of our project due to time constraints. Now we just want to make the markers small enough to be insignificant in size. However, I would prefer to just focus on finger detection with openCV. I dont know if the ZCam has the capability of accurately detecting really small markers. It cant seem to tell the difference between orange and red. Like I already said the colors are REALLY sensitive, especially to lighting changes.
hmm.. try a brighter orange and a redder red?
You may want to look into an HSV (hue saturation value) representation of the color image instead of the default RGB; it could help with some of that color detection stuff. As well you may want to look into disabling auto white balance. I had some similar issues last semester when working with the ZCam and that helped.
Yeah, computer vision is like that sometimes. Color detection sounds easy, but it isn't:
http://web.mit.edu/persci/people/adelson/checkershadow_illusion.html
Shadows - and lighting in general - are a big problem as far as color detection is concerned that humans mostly compensate for automatically.
Joshua Situka June 25th 2009
So we almost got the markers thing down..which is the easy part. I got the five movable appendage like fingers to work with different colors. Kelly is working on the making the appendages look like a hand..since we have to use simple shapes for the hands she is only using cylinders and spheres, and Mike makes fun of it, saying it looks like a blown up rubber glove. He's kinda right. ..and just found Shaq got traded to Cleveland, Vince to Orlando, Farrah Fawcett died, and that Michael Jackson just got hospitalized with cardiac arrest...there too much entertainment/sports news going on for me to concentrate. I'll pick this up tomorrow
Joshua Situka June 21st 2009
Father's Day came and went, and for some weird reason I'm at the lab telling myself to do work but it's just not happening. Eventually I'll decide to something today. It will probably be right before I go to bed and realize I needed to do something for Monday. Oh well. I'm hoping I can get this Hidden Markov model thing going for this week. Sounds cool....but really difficult. On another note, I finally finished reading at least 2 chapters of the book "The World is Flat" and I am now suppose to present the chapters to the class..At least that's the deal I have with Steve. Also, noticing that no one here goes by "Dr." even though they have their
PhD . I guess thats just the sign of the culture over here. if I ever got mine, would go by "The Dr. Tuka" the "The" is important because gives it a nice little ring.
update your blog!
How is "The World is Flat"?
What you should really do is pick up an MD,
PhD , and JD then you can be Dr. Dr. Dr. (although lawyers, JD's, don't actually go by doctor but it is a doctorate degree). There is an option at Iowa State to concurrently get your JD and
PhD (in partnership with Drake--in Des Moines).
Joshua Situka
So this openCV stuff is a little confusing but we finally got the book. im trying to decide what our next step is going to be once we get the cube rotating. This stuff is kinda hard working on teams because it seems they way things are going, one person cant do their job till another is finished with theirs. Or others are kinda waiting idlely because their task are sometimes not that tangible till the other person get done.
On the good note, now i'm pretty convince Linh is a genius who gets bored easily. He deals with complicated code very easily. He got the 3D cube rotating with our hands and nothing else between the time I started writing this blog, to when i finished sentence.
Now on to the hand.
Joshua Situka
Our C++ project is all of sudden really cool. Linh messed around with it for a day and a half now the thing looks and sounds amazing. I added some bells and whistles and Kelly did the last finishing touches. This programming with other people thing is kinda weird at first but then you get use to people displaying their different strengths on the keyboard. I'm hoping that our REAL project is going to go this smoothly. We actually have deadlines now of what need to do.
Rotate a cube with/without marks - 1 week
one finger rotation with markers - 2 weeks
Multiple finger rotation with markers - 1 weeks
Multiple fingers without markers - 2 weeks
We'll see if we have enough time to complete everything...and do papers/presentations/classes. (i'm really glad STATS is over.)
And this modeling things is really cool, how come they didnt teach me nothing but Texas History in high school/junior high/and elementary. This would have been a much better use of my time. I just spent it sleeping anyways.
As the summer goes on, I'm noticing my creativity towards my blogs are starting to suffer because im usually too tired to care about doing anything special. oh well.
that's good you have a plan
Thanks Josh. It feels like I've finally found an oasis after seven years of wandering through the desert. It sounds like you will really be using all the stuff we've been learning in our programming classes. The same goes for our group.
Joshua Situka
I kinda wished they told us how to do literature review and research questions in earlier crafts of research classes. It kinda helps us focus in more on our research becaause we are now forced to try and answer specific questions. Thank god the HCI course was cancelled yesterday, I just did not have the time I wanted to complete the work assigned the way I wanted.
I'm really tired of reading. I'm going to take a break today.
And now i just remembered that our group has to start on the big tic-tac-toe project.
There goes the weekend. I never even finished the last HW assignment for
OpenGL .
I'm gonna pay attention more to deadlines, them research assignments sneak up on you like a rattlesnake.
AmyGreen - 12 Jun 2009 - 15:15
NEGATIVE!?!?!?
Joshua Situka
I'm wondering why it takes me twice as long to write papers than to write code.
PJ Campbell - 11 Jun 2009 - 22:07
...because codes are logical and papers aren't!!!
IvanOjeda - 11 Jun 2009 - 22:05
It takes me twice as long to do both things compared to you guys >.>
it takes me twice as long to write code than to write papers
Joshua Situka
Our group got a little lucky this summer. Turns out the Z-cam experts Brett and Joe have a side project that goes hand and hand with our project. So they are helping us turn object colors into markers for the z-cam to detect. It's some really cool stuff.
It's cool that people in the lab are pretty helpful.
Thanks for the birthday wishes. It always nice when someone else has done some of the work for you. The Digital Games group project is based on this research group in Switzerland. They wrote a paper that explains exactly how to add logging capability to digital games. It's very helpful. Yea, when Kareem was at UCLA they banned dunking because he was so dominant. Much to their dismay, he developed the sky hook and continued to dominate.
Joshua Situka
Music, bells, whistles, and the Mario theme song and our group still didnt win the game presentation. oh well. Eric didnt even have a prize so i guess its just for VRAC bragging rights. But Vijay said he has something ready if we win the opengl competition. Basically now, instead of the hand we are going to put tic tac toe on steroids, and make it look really cool. Apparently the hand takes ALOT of work.
Progress on big project is still not really going anywhere. Still, just need more time I guess. Joe and Brett have given us really good info but today didnt get really any free time till now and its 4pm. So prolly put in some later hours again. I will start to refer to my caffeine schedule (please reference June 3, 2009).
come to the scavenger hunt
Tic-Tac-Toe on steroids sounds to me like you're going to implement Tic-Tac-Toe Cube (or 3D or whatever you want to call the 3x3x3 version of Tic-Tac-Toe). Some animation effects for winning would be nice too. Maybe some random 'fog' to obscure the board at times and make the game more challenging. Be creative with it
Joshua Situka
I'm really looking forward to messing around with the opengl stuff. Our group is going to create a Mario type hand which can throw fireballs. Well at least that's what we plan on doing. It coincides with our project because Song wanted us to create a hand anyways with the z-cam. I'm hoping we get a lot done this week.
We start STAT today. The STAT and HCI courses are at odd times like 3:30pm-4:30pm. I usually dont get too much accomplished in the 30 minutes after. I rather they just be right after blog so we can get them over with and work on our project straight through.
PJ Campbell - 09 Jun 2009 - 15:16
I'm happy to report that I've posted today...
BTW...please resize those photos
I can see if I can switch the HCI course a little bit at least. It would be easier to change it from 3:30-5 (it's an hour and a half) then right after blogs--would that still work alright (you should still get the straight block of time for research, and that way if you all finish blogging earlier you don't have a weird block of time to waste you can start working right away).
me too
Joshua Situka
After talking with Song about the project turns out that Microsoft only uses joint recognition and what we are doing is much more full detail on the body part. So I feel much better about the project, now just waiting on the installation of HTK on our computers so we can start playing with the Sign Language software.
And I'm planning on doing some real cool stuff to our tic-tac-toe project. I want to get music playing when a player wins and have some failure song if there is a tie. I want to win the competition.
My mentor makes these faces that are really goofy. And randomly does karate. And has some type of machine that supposedly senses radiation, (and for some reason he felt that he should test and see if our gum drops had radiation). This stuff is pretty funny. I didn't imagine this as a research lab when I was a kid. I thought everyone will be wearing white coats with only two words in their vocab "glyvn" and "glavin". Alot of these people have interesting personalities.
Funny how your blog reminds me of "mad scientists" or "the absent minded professor"...
IvanOjeda - 05 Jun 2009 - 20:34
LOL. I hope you can put some awesome music on your game. I'm expecting some salsa.
hah! if there is anything the Simpsons taught me about nerdy scientists, glyvn glavin is one of them
There's definitely never a dull moment with Kenny.
Joshua Situka
We pretty much finished our tic-tac-toe project. The only thing to do is to comment the program, make it user friendly, and erroneous input protected. However, our real big project is interesting. There is this sign language code that right now doesnt make too much since, but that's because im more rusty in programming than how the code was written. And also it turns out that X-Box just released some huge Z-cam thing on monday that can already do what we were working on. Now I feel that the wheel was just invented in another part of the country, when were just figuring out that making some round might be a good idea. (I could make a better analogy but whatever). We have a meeting with our faculty advisor and wonder if this new development is going to change our project tasks.
Josh, if you talk with the folks who have done zcam work around here, you'll hear them complain about its bugginess. This problem is by no means solved; that video is just Microsoft marketing. Song knows that if this research weren't needed (and a hot topic), he wouldn't be getting funding and have you doing it. By the end of the summer, you'll probably have a better video.
Hey, Josh... Don't worry about "re-doing." It is quite possible that you guys went about the problem in a different way. Some of the advances in science have come about this way. Congratulations in finishing up!!
Erroneous input is the problem of the erroneous user.
that's good that your tic tac toe project is about done, I don't totally get your analogy but I think I know what you're trying to say.
Okay, im starting to realize that there is alot more stuff due than what I expected. I am not worried about getting anything done, but I am little worried about getting it done to my professor's and my own standards. I am a little bit of a perfectionist when I do care about something. And making sure all my programming, work, reading, and NSBE stuff at home is properly taken care of, is probably going to be real tiring. On the other hand, this past spring semester has already prepared me for any sleepless nights resulting in caffeine overloads on the body. I usually have to cycle my caffeine sources for them to take any effect. It's kinda like how body builders have to cycle workouts so they don't hit a plateau. I think this summer I'll go with this schedule.
| week |
Caffeine Beverage |
| 1 |
coke |
| 2 |
green vault |
| 3 |
reg coffee |
| 4 |
monster |
| 5 |
redbull |
| 6 |
5 hour energy shot |
| 7 |
red vault |
| 8 |
vanilla coke |
| 9 |
turbo coffee from burger king |
I recommend the 5 hour energy for week 10 (unless you count this week as week 1) as there'll probably be at least one all nighter... I have a 12-pot coffee maker here and I'm happy to make extra coffee if there's interest in it.
Joshua Situka 2 June 2 2009
This lecture today with Stephen over cognitive/tutor learning and Carnegie Learning software was really cool. I didn't realize there was research that much in line with what I wanted to do. I feel that is something that I could really contribute to and is the reason I decided to get into the HCI field. I'm now really glad I took this research opportunity over the fun summer in Chicago making GE money.
I have a paper that takes a critical stance to intelligent tutoring--I don't agree with it, per se, but I think it's an interesting read and has some ideas worth thinking about. If you're interested in it then let me know, and I'll send it your way.
I didn't know you haven't gone yet, next time I go, I'll give you a call.
that's great! I didn't know you had an interest in software for learning.
Too bad we didn't have this technology when we were in high school, huh?
Joshua Situka 1 June 2009
- pics_155.JPG:
This is a light and its intended use is to light up rooms with light so you can see.
However, its usability suffers because of safety. If you were trying to replace the bulb inside you will need at least one other person helping you
because of the size of the light. Even then it would be awkward to handle and give to someone else because of the size. What would be better is to have a light that could be handled easily with one hand so that it could safely be replaced.
- Controllers:
The Nintendo Game Cube controller on the top is intended for you to play video games interactively on a television. However, its usability suffer's because it is inconsistent with many controller designs of the past. Many contollers (as seen in the X-Box 360 below) have more of a handle for you to use and the button layout is more symmetrical. Therefore, adjusting to this controller scheme is awkward for no apparent reason. They actually came out with controller's after this one that address that issues with more handle.
Joshua Situka -29 May 2009
So CAVS lost, which is unfortunate. But at least I can spend more time doing my research. We are probably going to do a Tic-Tack-Toe game for our C++ project, we only have a week to do it. I feel real confident about the projects that I am doing, And I still cant understand how I tested out of the programming test, i guess im an excellent guesser, because I didnt know half that stuff. Oh well. I just want to start playing with the Z-cam, and downloading
OpenGL , and Matlab to my laptop to do the work at home. I feel like i'm rambling so im going to stop.
Main.Joshua Situka -29 May 2009
I actually know what we are suppose to do now with the research. I was EXTREMELY lost at first, but with alot of work I think our team is going to do a good job. Apparently, Markerless Motion Tracking is going to be the craze in Hollywood and Video Game Design as soon as the find out an easy way to integrate the technology with the games and movies.
Also, the
LebrONS won! Yeah. Just a two wins away from my dream final...if the Kobe's win tonite.
Please add the comments code to each blog entry so people can respond to your posts (just copy and paste the code block).
--
JoshuaSituka - 28 May 2009
I never thought I would ever blog because I use to make fun of people who did when I was younger. This is karma. So I have been in Iowa for 2 days now, and everything is going pretty well. I like the diversity of the people in the REU program and think I will have a lot of fun this summer. I am not really looking forward to hiking this evening, I rather play basketball or watch the CAVS, but I will try to make the best of it.