In light of the fact that I know my motion tracking needs a lot of work, I decided to start trying different methods for getting and working with information from my webcam.
Last night (and a little over lunch today) I threw a couple quick experiments together to see what it would look like, and I think that I might try this out with the video when I got home tonight.
The first example comes in the form of a basic color picker that lets you select the color of a pixel from within the image (on the top left). Once you have selected a color, use the slider on the right to control the threshold.
The second example is very similar to the first, it just uses a different sampling method for the color that it is going to use with the threshold. To try this tool, draw a box (from top left to bottom right) over the region of the image where you would like to get your sampling from.
This was a very quick and dirty experiment to see what I could get out of a bitmap with some basic functions, so please forgive my source. I will post more once I get it cleaned up tonight and start trying to work it into my motion detection application.
A couple of days ago I posted my first attempt at motion tracking in AS3 using a webcam and promised that I would play a little bit more with it. Well, I got a little bit anxious and before refining the tracking code I decided to put it to use.
You may have seen the video of head tracking using the Nintendo Wii. That video inspired me to try my hand at the same concept using Flash, Papervision 3d, and my newly discovered love for trying to get motion tracking / facial recognition into Flash.
As such, I have created this example which is my first step in creating some sort of tracking interaction with a 3d scene. if you would care to see a little bit more detail about what is going on, hit X to see the video and W to see the framerate. Stay tuned for more updates as I update the tracking code. Once again, I will post the source when it is a little further along.
** please note that you require a webcam to take advantage of the motion tracking, if you do not have a camera the scene’s camera is simply controlled by the mouse.
For the past 6 months or so I have been hearing the term “Devigner” or “Hybrid” floating around the Flash industry. While having many talents may be a wonderful thing in small-scale agencies and freelance work, I don’t believe that the skills of a Designer and a Developer should be looked for in the same individual.
I started out in the industry as a graphic designer and web developer but quickly came to realize that the more code I wrote, the less design prowess I had. The trend in the industry to find a hybrid developer / designer is a good idea on the surface but has many flaws. I have always been a fan of the phrase “jack of all trades, master of none”. While it is a great asset to know a little bit about what is happening on the other side of the fence, it should stay on the other side of the fence.
Both designers and developers are creative and skillful in their own ways so why should they try to do it all? I have worked on countless projects where I have been put on a project that was “almost completed” by a designer. Upon getting the files, I am told that I just have to do a little bit of optimization and get the files delivered ASAP. More often than not, once I take a look at the files it is obvious that they need a LOT of work. The designers that I work with are spectacular at what they do, but when they try writing code I am given files that take hours just to understand.
If you would like to be a developer, take a course or two and practice. I didn’t learn to write code over night, nor do I try and create complex designs expecting a designer to pick it up and think it was a job well done. There is a lot more than just being able to write a frame script in Flash that goes into being a good developer, especially when you are creating international pieces.
When I am busy and on a tight deadline, I don’t want to (nor do I have the time to) spend hour after hour ripping apart a project just to re-write 90% of the code that went into the “almost completed” version.
An ideal situation for me is to be able to sit down with a designer as they are starting to work on a project and come up with boundaries and share ideas. I respect their skills and would like to utilize them as much as possible. Just as it isn’t efficient for me to rewrite an application from them, it isn’t efficient for them to spend days creating things that could have been automated with a little bit of code. It is also inefficient for them to realize that it can be done with code and try to do it on their own.
So next time you are thinking about becoming a hybrid or hiring one, take into account the fact that it isn’t as easy or productive as you might think.
I have been doing some recent experiments in ActionScript 3 so I thought that I would start posting them. Most of my recent work has been client work using Papervision, but I found a little time to play with my webcam over the weekend.
I have a couple of uses planned for this and will post them when they are done but here is the first iteration.
With the help of a diff equation from Forrest Briggs, I managed to capture the specific area of movement within the webcam frame. I will continue working and post source once it is complete. This little project is helping me understand the workings of the bitmap data objects in as3.
Flash in the Can Toronto 2008 has come to a close and it is back to the real world. I had a great time at the conference and got so much inspiration from what I saw. As always, the presentations showed that there are some people truly pushing the limits of Flash and creating some beautiful things. I got to attend all of the presentations that I wanted to which included most of the technical track.
Yesterday, Grant Skinner talked about his inspirations that drive his work. It was great to see what others do to keep their interests up when they are held back by corporate site limitations. I also managed to catch Keith Peters‘ talk about Chaos which showed how crazy the world of math is.
It seemed as though the focus of many developers in the conference was randomness, with a side of Perlin Noise. It was interesting to see the different methods that everyone uses to generate their work, and it was especially interesting to see just how much randomness their is in the works that look so precise and structured.
If you are interested in some info about the conference and what was talked about, the new FITC blog is a good resource. Most of the presenters also mentioned putting their slides online so keep an eye on their blogs.