Creating animated movie from digital stills?

debbie wrote on 2/6/2003, 1:01 PM
Hi,
I am trying to create an animated digital movie with elementary school students. The kids have created paper cut out figures with hinged body parts that can be moved. We are taking digital still pictures with a Nikon Coolpix camera and moving the body parts of the characters between shots. Then we are importing the digital stills into Vegas Video 3.0 to create a video track. We are planning to add an audio track with sound effects and kids' narration and background music, too.

Here are my problems!!
1. I have set the preferences for still shots to 3 frames because that should result in each image being shown for 1/10 of a second if the movie is shown at 29 frames/sec. However, something is not quite right because both the preview and the rendered file are much slower than this--you can see the individual frames as they change.

2. The pictures flicker because of changes in background color,etc. I am using 2 photographic lights of 250 watts set up at 45 degree angles to the kids' artwork. Any ideas to make this better?

3. I called the tech support person very nice) and he suggested that I could pull all of the pictures in a sequence into the video track at once by checking the still sequence box. I did this, but then the pictures seem to show much like the default still image time of 5 seconds.

3. I need to tell you that I am VERy new to all of this--like I've done the tutorial and no other digital editing at all!!! Please help me by explaining things in very easy terms or I won't get it at all. I hope that I'm not too frustrating, but I'm at home sick and the computer is at school and I'm just sitting here trying to figure it all out by reading the forum board, etc.

Thanks,
Debbie

Comments

philfort wrote on 2/6/2003, 1:11 PM
Your first #3 is a bug in Vegas. You can workaround it by "doing something" to the clip that results from the sequence in the clip bin. For example, right click it, and click on properties, change some stuff, change it back, and hit ok, and then the images should last for the still shots length you set in PReferences.
Chienworks wrote on 2/6/2003, 1:16 PM
Debbie, try this ...

To begin with, make sure all your still pictures are named with sequential numbers, for example,
frame 001.jpg, frame 002.jpg, frame 003.jpg, frame 004.jpg ... etc. with no gaps

- Click File / Open, browse to the directory where the stills are stored and click on the first one
- Check the "Open still image sequence" option
- Click OK
- A properties window appears,
- Change the Frame rate to 10. There's no selection for this so you'll have to type it in manually
- Click OK

You should now have an event on the timeline containing all the pictures that runs at 10 pictures per second. Beware that if you have any gaps in the numbering sequence this process will stop at the gap and you'll have to open the next set as another sequence.
Chienworks wrote on 2/6/2003, 1:25 PM
Is there any other lighting besides the two 250W lamps hitting the figure? If so, then people moving around in the room can change the brightness and character of the illumination quite a bit. Turn off all other lights, cover the windows, make sure people aren't moving around near the figure. This will help the consistency quite a bit.
debbie wrote on 2/6/2003, 1:31 PM
Hi,
Thanks for the suggestions on the digital still sequences. The camera numbers the images sequentially, so that should be ok.

I am not using the flash on the camera, but there are overhead lights and kids walking around. I'll experiment with turning the lights off tomorrow.

Thanks,
Deb
debbie wrote on 2/6/2003, 1:32 PM
Thanks,
I'll try this tomorrow!

Deb
SonyDennis wrote on 2/6/2003, 2:21 PM
As others have already mentioned, I will re-iterate here: there are two ways of doing this: each still is an event, and you make really small events (using the 'default still length' pref), OR bring in the images as a 'still image sequence' where they will be treated like a video (where each frame is sourced from a still image) which can be edited in the normal ways (trimmed, FX applied, etc). If the duration for each still needs to vary, the first method works best. If the cadence is steady, the second will automate a lot of the manual work for you.

By the way, small variations in lighting and model positions can sometimes add the the overall look of your piece, so don't sweat them too much.

I'm glad to see teachers doing these kinds of things with their kids.

I wish I still had the stop-motion video I shot with my daughter and her beanie babies, it was a riot.

///d@
debbie wrote on 2/6/2003, 2:50 PM
Dennis,
The cadence of the film needs to be consistent, so the second method that you suggest seems like it would be easier. We just finished a practice film called TAG's Magic Hat, where stuff comes out of a magic hat. Each student created something for the hat. Anyway, we imported the pictures one at a time for that film. Now, we're going to do the real thing--a film about Louisiana's historic places.

As I have no idea of what I'm doing, is there any special kind of format that I should choose, etc. for this project? Hopefully, we'll get someone at the school board to post the project on the school's webpage. Is there anything that I should know about this?

I wanted to make a CD for each participating child and through reading the forum topics, it seems like I need special software to burn CD's with the project. Is this correct?

Thanks for the help. I'm in way over my head on this learning curve, but it's fascinating and the kids are loving it. They're actually using the software. I'm just staying a step ahead of them and trying to troubleshoot. And I'm frazzled about it.

Debbie
dmcmeans wrote on 2/6/2003, 8:14 PM
Debbie,

>I wanted to make a CD for each participating child and through reading the forum topics, it seems like I need special software to burn CD's with the project. Is this correct?

If you're wanting them to be able to watch the video on their home DVD player, then you're talking about making a Super Video CD (SVCD). Most DVD authoring programs that you can purchase will also make SVCD's. If you want to skip the extra costs, you can use VCDEasy (www.vcdeasy.com) to author the SVCD disc. It's free.

Obviously, the caveat with SVCD's is that they don't play in all DVD players.

But if you just want them to be able to watch the movie on their computer, then a Windows Media Video (.wmv) version will work just fine. (I'd use the 3Mbps template, tho').

I hope you are able to post this video. My kids would get a kick out of watching it.

David
SonyDennis wrote on 2/6/2003, 8:49 PM
Debbie:

I'd suggest using the "Multimedia" project template, it won't be too demanding on your computer, and matches the output resolution for web video.

You can burn VideoCD discs right from Vegas, and these will play on many DVD players. You can also play them on PC's, although sometimes they need software installed to do this. In that case, Windows Media Video (WMV), MPEG-1, or QuickTime are all good solutions, just put the data file on a CD for delivery. You've got lots of help here in the forums from people who support the idea of using video and technology in education.

///d@
TomG wrote on 2/6/2003, 8:55 PM
David,

I tried to go to www.vcdeasy.com and it wasn't found. You sure that's the right address?

TomG
wcoxe1 wrote on 2/6/2003, 9:57 PM
I can easily account for lighting changes. As mentioned, changes in the room light, time of day, people moving about, all that can change things. But one thing that is almost always overlooked is the change from picture to picture, itself. Changing ANY of these things can change the amount or color of the light. But think of THIS:

Small changes in the amount of color IN the scene, such as moving something, or adding a different character, or taking one out, can easily fool the automatic meter on the camera into thinking the scene is lighter or darker.

The way to accomodate (avoid) these changes is actually very simple. Go to a good camera store and buy an 18% Grey Card. Keep this card VERY clean. Place the card over a scene such that it completely fills the viewfinder. Take a meter reading of the gray side (not the white, if there is one), LOCK the exposure reading, and shoot. You see, a background, by itself reflects one amount of light, but a figure in that background reflects a DIFFERENT amount. Changing the figure, adding, or subtracting a figure, changes the AMOUNT of light. The gray card is the professional's way of getting a true AVERAGE. One that the camera is actually tuned for, in the first place.

If ANYthing in the room changes, people moving about, sun moving because of time, etc., different light turned on/off, you need to read the gray card, again, and lock it down, again. Works well for changes in the AMOUNT of light seen by the meter.

However, the gray card can't help much with COLOR changes of the light, such as turning on or off a different light, or the sun changing, or going behind a cloud. Try to get rid of all the variables and the card trick will work best.
VideoDentist wrote on 2/6/2003, 11:12 PM


Can these animated movies from digital stills be created from my mini DV camera (Cannon GL2) instead of a still digital camera. ie. Could I video a still image in short bursts then appl them to the timeline with good effects?
TorS wrote on 2/7/2003, 1:58 AM
Any series of images that you can load into your computer can be used that way. There might be an optimum as far as sizes and resolutions are concerned, but I think Vegas will eat it anyway, and render it to whatever end format you want.

Are you sure you can't do video animation recording (step recording) with your Canon?

Tor
mikkie wrote on 2/7/2003, 8:39 AM
FWIW, I think the closer to the frame size the pic is, the less image manipulation the software has to do - on the other hand with video requireing less resolution then still photography doubt it would matter all that much.

Just about any technique will work - think of any variances as added personality. besides using cameras, I have heard of people using scanners, & webcams are popular. As far as light goes, you want something white (versus halogens or flourescents). Softer (reflected) light might work better to avoid shadows & reflections, not show handprints, glue and stuff as much. Besides bounceing the light off the ceilings, a few poor man's alternatives can be made using cheap aluminum window screen frames -> using those clips that hold a stack of papers together, or the one's that hold a bag of chips closed, cover the frame with thin white material like from a bed sheet and shine the light through for a light box effect - cover it with something colored for a diffused gel - something reflective or at least thicker to bounce &/or fill.

Digital cameras/cancorders often try to automate everything including the color or white balance. IF you can't set it just once, one trick that's neat with digital cameras is to have the white card handy (or simply a sheet of bright white inkjet paper?) -> most digital cameras have some way to lock the exposure settings (with cheaper cameras you just keep holding the shutter button part way down) so place the white card/paper over the subject, lock exposure & white balance, and pulling the paper away, shoot.

One thing that might help sometimes is pre-processing any photos. Most image editing software has batch capabilities built in, as well as automatic image/color/level adjustments. While these auto adjustments may not suit everyone's tastes, they are consistant, and might bring any differences between frames closer.

A lot of (most?) animation is done at something like 12 frames a second, then frame doubled to 24, then telecined to 29.97. I think most of the stuff talked about in this thread get to the same place through various routes, so not adding anything really - just in case it helps anyone conceptualize how they'll try it.

mike
debbie wrote on 2/7/2003, 8:55 AM
Hi VideoDentist
When I was researching this project, I found out that DV camcorders with single picture capacity actually records about 140 frames for a single picture. I decided to use a digital still camera because I didn't know if Vegas 3.0 could compress all of those images and I didn't want to have to do it picture by picture.

If Vegas could somehow take all of the camcorder images and either compress them or run them at a extremely accelerated rate, it seems like the animation would work.

From what I've read, animation must be shown at least 10 images/second for the human eye to perceive the movement as animation due to persistence of vision. Anything slower will look like a slide show.

Is there a difference in file size or type when using a digital camcorder to take pictures as opposed to a digital still camera? This takes up gobs of space!

Thanks for all of your help!!! Ya'll are wonderful people.

DEbbie
bakerja wrote on 2/7/2003, 9:02 AM
Maybe I missed this but, I didn't notice anyone recommending locking down the exposure and color balance. Most cameras have automatic settings enabled by default, but you can ususally set to manual to maintain color balance and Iris.

JAB
mikkie wrote on 2/7/2003, 9:50 AM
Yeah I did JAB, & so did wcoxe1 ;?b

"If Vegas could somehow take all of the camcorder images and either compress them or run them at a extremely accelerated rate, it seems like the animation would work."

If each *still* took up the same amount of frames, then making things run faster is easy - right click, properties, rate - though not sure the amount of total work would be any easier.

Thinking about it a bit more, wonder if something like this company's cameras might not work: http://www.aiptek.com/

They've got quite a range of still cameras that while lower resolution also can be used at reduced frame rates as webcams, yet still have slightly better optics. Idea would be to set one up on a tripod or similar, hook it up to a PC or laptop, it would show the image on the screen, and you'd click to save the image. Same final result as everything else, but you save the step of having to import the still into your PC, and most webcams can lock exposure etc. Anyway, it's a much cheaper version of what I've seen at the DMV for driver's license photos.

mike


VideoDentist wrote on 2/7/2003, 11:09 AM


Just a little step back. To make an animated scene of clay models for example, how many photos would I take of each situation before I change the position of the clay model? I would like to try this.
debbie wrote on 2/7/2003, 12:11 PM
We've been taking one shot in each position. I did claymation with a Super 8 camera and film several years ago with my students. Be careful! The lights melted the clay! Use plasticine--Plasalina is the brand name that we have. It's loads of fun, though.
wcoxe1 wrote on 2/7/2003, 1:27 PM
If you want to stick with the camcorder, the one that takes 140 frames, all you have to do is let Vegas capture it, and then CTRL+ALT+-> (right arrow) to jump to the beginning of each clip. Then, expand the zoom (only the first time) so that when you just press -> you move ONE frame. Press S to split, click on the BIG part, and press delete. That will leave exactly ONE clip. Repeat.

I don't know enough about programming, but I'd bet that the new V4 scripting could automate all this.

What do you say, guys and gals, who is going to be the FIRST to post that script? And, who will do it with the most elegance?
VideoDentist wrote on 2/7/2003, 1:27 PM
Thanks. I wasn't sure if one shot per movement was correct. I guess the movements between shots should be small?
Chienworks wrote on 2/7/2003, 2:53 PM
It wouldn't really make sense to take more than one shot between movement, since the individual shots could simply be lengthened instead. I used to do tons of these as a kid with 8mm and it took quite a bit of practice to get the timing right. The hard part about practicing was that it took two weeks to get the reel of film back from the processing company, and my 10 year old allowance didn't afford me film and processing more often than every other month anyway ;) So i spent the first couple of rolls just experimenting with timing, lighting, etc.

PTL for video!

Here's a couple of quick Lego building samples i did last year. Sorry that one of them is RealMedia. Oops.
http://www.chienworks.com/media/lego001.mpg
http://www.chienworks.com/media/lego002.rm

I'll admit i captured these with Ulead's Media Studio. It has a very nifty capture function that adds the current image as the next frame in a continous .AVI file each time you press the space bar. *sigh* If only SonicFoundry had this ...

I did do a project with sequential stills taken with a digital camera here:
http://www.vegasusers.com/vidshare/textdisp?chienworks-nightwalk.txt

These were imported as stills in VideoFactory, grouped, then stretched (shrunk, actually) to fit the beat of the music.

I think i'm gonna try a Lego castle tonight.

ok, ok ... so i'm feeling 10 years old again ;)
debbie wrote on 2/7/2003, 3:19 PM
The smaller the movements, the more life like the animation. WE move the cut out characters about 1/4 inch. Of course, bigger changes translate into faster movements--because speed = distance/time. If the kids want the character to move quickly in the animation, they make the changes bigger. Smiles..
TheHappyFriar wrote on 2/7/2003, 3:32 PM
How big are your figures? Back in college we put a piece of plastic over a scaner, and would set out various figures in different poses to be scanned in at once (in some type of order). Then we would goto in photoshop and split image into the frames we needed. The light would be conistant for all of the images because they were all scanned at once. If you can scan them, you could probely get what you want a little faster then using a camera. Just make sure you scan everything at the same DPI and dimensions, and use a grid in photoshop.