Kevin's “pictureMagic” Page

The images shown on this page are presented using a Java Applet which sequences thru a set of pictures, featuring a variety of “image transformations” as it goes from one picture to the next — much like we're all used to on the TV News and Weather. Click here for further details.

The images shown above are presented using a Java Applet for which the user interface is detailed above. I separated the Applet screen space from these details simply so that the pictures appear at the top of the page. The Large (640 x 480) picture — especially — forces all the text except that in the Table to get pushed below the bottom on the initial screenfull.

I am beginning to suspect that when the Applet initializes, the "previous image" (for the purpose of the transformations discussion) start out being a completely black plane the size of the entire image space. Until I come up with better names for each of these transformations, they are — in the order presented — called:

• “brushfxmod” — a solid black background is drawn, but you can see that the destination image is behind that as soon as a large brush begins moving over the image. And as it moves, the underlying image appears more and moe as though the brush were clearing away they the blackness. The motion is either horizontal left-to-right swipes, or it's circular working from the inside out. I don't know how it decides which of these motion paradigms to apply.

• “zoomfxmod” — does a Zoom In, focusing on the center of the image and going in until just a few pixels fill the entire image space. This one is 'blind' in that it's digital zoom is purely algorithmic. Witness the fact that when you start with the "previous image" being the default initial image (all black), it appears to just stay all black because that's what Zooming In to any solid color gets you. Once that transformation has finished, assuming that the real destination image is an actual picture, then you see the Zoom In happening again but in this case it's really the 2nd iteration of the sequence in which case you'll be getting to a different destination image.

• “blackholefxmod” — The "previous image" is rolled top away from the viewer wrt a horizontal axis defined to be thru the mid-point of each of the vertical sides of the image. This gives the impression that the top of the image is moving away while the bottom of it is Zooming towards you. When the 45 degree angle point is reached, the flat plane surface transforms into a cone shape rotating counter clockwise and getting smaller (tighter) like a funnel cloud in a toronado. The funnel gets smaller and smaller until it disappears as a vertial line.

• “unseenwindsfxmod” — the "previous image" as a flat plane gets a little smaller so you can see it as a separate entity, and then it moves like a flag blown in the wind. The flag then does a horizontal roll away from the viewer, and finally is rotated clockwise until it is "flown" out of the image space in one of the corners. I don't know how (randomly?) it decides which of the four corners to exit via.

• “origamifxmod” — the flat image plane is slowly folded up, reminiscent of an origami demonstration, into a folded-paper airplane which then flies in a horizontal circle counter clockwise, exiting via the left side of the image.

• “liquidfxmod” — This one is like brush in that it progressively lets more and more of the destination image "show thru" relative to the "previous image", initially black, that it starts with. This one is very dependent on having the "from-" and "to-" images be different because if this is not the case, you can hardly see that anything is happening save for a warping of the image in various blotches that seem to move around.

The only reason I have 8 images but just 6 types of image transformations is because I wanted to experiment with what parameters I can pass to the Applet that were not used in the model that this page is based on. To that end, I re-used “zoomfxmod” and then “origamifxmod” for the 'extra' two pictures that I couldn't resist adding. Since this is a technology experiment... the content is irrelevant and so the pictures themselves are simply 'randomly chosen' snapshots of my family, primarily because they are the ones most likely to be impressed by this notion of "TV on the Web".

For now, this “sample” is not on part of my Java Applets Portfolio because I didn't write the Java code myself. Rather, this “exercise” demonstrates a typical way to create a highly-parameterized mechanism which runs inside a browser but essentially does “everything” in Java via an Applet.

If you have any questions about this mechanism, just ask me or look at the model that this page is based on.

Pammett Family Home page