i will be drawing them. an interesting idea, but i think it should be easier than that. if i use a standard size paper, proper scanner setup, and proper alignment i should be able to process the cropping and saving of each cell with some kind of photo manipulation program. i did something like this once with photoshop where i took a batch of static images, cropped into 4 quarters and saved each quadrant as a series of separate images to be animated. this would just have to crop out a series of ~250x160 px rectangles and save each as a numbered image for animation. i'm mostly concerned with proper alignment so that the crops will be exact, ... well roughly enough. i don't mind a bit of imperfection entering here, the idea is to capture the human essence in quick animation.
On Fri, Jun 3, 2011 at 8:17 PM, Tyler Leavitt <[email protected]>wrote: > Are you drawing the animations? Could you create a template with all the > cell numbers in the right place and print those out and draw on that? Then > when you scan it in, use the Tesseract library ( > http://code.google.com/p/tesseract-ocr/) in some way (I've never used C > before but maybe you have). From the limited reading I've done, it appears > the Tesseract program outputs the text in an image file into a text file. > That doesn't sound like it could help you, but if you can program that > sounds like the library you'd use. > > Tyler > > On Fri, Jun 3, 2011 at 4:51 PM, grant centauri <[email protected]>wrote: > >> now that i think about it more, perhaps a physical solution for >> registration would be best... like taping animation pegs to my scanner or >> something. i may be making this too hard on myself. ;) >> >> On Fri, Jun 3, 2011 at 6:40 PM, grant centauri <[email protected]>wrote: >> >>> just a question for the community. >>> >>> my idea is this: >>> >>> i'd like to replicate the experience of animating directly onto 16mm or >>> 35mm film by using some kind of "exposure sheet" which can be scanned in and >>> automatically chopped into frames and optionally converted into video. >>> >>> the sheet would have rows of "cells" representing the frames of film >>> which could be drawn on and perhaps some kind of registration marks for the >>> processing? >>> >>> does anyone know of any tool that can be used to detect registration >>> marks and then perform the cropping necessary to get each cell into its own >>> image file for animation? i imagine that imagemagick could probably do >>> this, the thing i'm mostly concerned with is registration. i'd guess it >>> would be difficult to get each scan perfectly aligned, but if there was some >>> kind of registration marks maybe the computer could align them with a >>> 'cropping template' somehow. >>> >>> i'm guessing this shouldn't be too difficult, i may have to do some >>> hacking, but i was just wondering if anyone out there had any leads i could >>> follow. >>> >>> also, if there's a way i can filter out the scanned background to emulate >>> clear film that would be great too. perhaps this is in vain. >>> >> >> >> --- >> [email protected] >> http://identi.ca/group/puredyne >> irc://irc.goto10.org/puredyne >> > > > --- > [email protected] > http://identi.ca/group/puredyne > irc://irc.goto10.org/puredyne >
--- [email protected] http://identi.ca/group/puredyne irc://irc.goto10.org/puredyne
