mirror of https://github.com/python/cpython
a43111118f
|
||
|---|---|---|
| .. | ||
| .cvsignore | ||
| Makefile | ||
| README | ||
| VFile.py | ||
| Vedit.py | ||
| VeditForm.fd | ||
| Viewer.py | ||
| Vinfo.py | ||
| Vplay.py | ||
| Vrec.py | ||
| Vtime.py | ||
| cam.py | ||
| camcorder.py | ||
| cmif-film.ms | ||
| colorsys.py | ||
| i2v.c | ||
| makemovie.py | ||
| squash.c | ||
| squash2.c | ||
| statit.py | ||
| syncaudio.py | ||
| tomono.c | ||
| tv.py | ||
| v2i.c | ||
| vcopy.py | ||
| video.py | ||
| vinfo.py | ||
| vpregs.py | ||
| vtime.py | ||
README
CMIF video tools ================ This directory contains Python and C programs to manipulate files containing digitized video in the "CMIF video format". History ------- We started this in October 1991, when we had a large framegrabber board on loan from SGI for a few weeks: we developed a movie recording program (camcorder.py) and added numerous features, including still frame capture and synchronous sound recording using a second machine (the machine holding the framegrabber board didn't have audio). During the following months, when we started using and processing the recorded film fragments, the "CMIF video format" was revised several times, and we eventually created an object-oriented interface for reading and writing various incarnations of these files, called VFile. (This module is also used by our flagship application, the CMIF editor, not in this directory but in /ufs/guido/mm/.) When we got our own Indigo entry-level video board (in June 1992) and a version of the Irix video library that supported capturing PAL format (in August 1992), Sjoerd added an interface to the video library to Python (sv) and Guido wrote Vrec.py (based upon a still frame grabber by Sjoerd, in turn based upon SGI demo code in C) to record a movie using it. Guido van Rossum Jack Jansen Sjoerd Mullender Overview of files ----------------- cam.py network real-time tv broadcast; see tv.py usage: cam [packfactor [host]] specifying 'all' for host broadcasts camcorder.py record video movies or make snapshots (in movie format) usage: camcorder [-c] [-a audiohost [-s]] [-p pf] [moviefile] -c color (equivalent to -p 0) -a audiohost syncaudio is running on audiohost -s start syncaudio (on audiohost) -p packfactor packfactor (default 2; 0 records color) moviefile (default film.video) keyboard commands: s stop grabbing (single step if already stopped) c continuous grabbing r start recording p pause recording (record single frame if paused) ESC quit Vrec.py record video movies using the Indigo video library and board colorsys.py color conversions, used by video statit.py various statistics operations on movie files syncaudio.py record audio synchronized with camcorder -a usage: syncaudio videohost soundfile soundfile format: 16 bits/sample, 16000 samples/sec, 1 channel (actually, you'd better set the sampling rate to 16000 with the audio panel) tv.py receiver for transmissions from cam.py vcopy.py selectively write frames from one movie file to another usage: vcopy infile outfile commands: 'n' gets next frame; 'w' writes current frame video.py player for movies recorded by camcorder.py usage: video [-l] [-p pf] [-m mag] [-F] [moviefile [soundfile [skipbytes]]] -p pf override packfactor (to zoom) -l looping -- restart movie when done -m mag magnification -F run at max speed moviefile default is film.video soundfile default is none (no sound) skipbytes byte offset in soundfile where sound starts Vplay.py similar but more modern, using VFile.py vinfo.py print summary of movie file(s) usage: vinfo [-d] moviefile ... -d print delta times (default: print abs times) Vinfo.py similar but more modern, using VFile.py vpregs.py definition of VP registers vtime.py virtual time module imported by syncaudio.py and camcorder.py Vtime.py (unrelated to vtime!!!) Copy a video file, manipulating the time codes (e.g. faster/slower, or regenerate time codes, or drop frames too close apart) Vedit.py interactive video editing program Viewer.py two viewer classes used by Vedit The following are C programs, either for efficiency or because they need to link with a C library: squash.c make a movie smaller by averaging pixels usage: squash factor [bits] <moviefile >newmoviefile factor x and y compression factor bits #bits left per sample in result (default 8) squash2.c make a movie smaller by dropping pixels usage: squash2 factor <moviefile >newmoviefile factor x and y compression factor tomono.c like squash2 but outputs a monochrome movie v2i.c convert the first frame of a movie file to SGI .rgb format link with -limage i2v.c convert an rgb file to "lrectwrite" format (this was used one time by the CMIF editor)