CorePNG v0.8.2

Jory Stone's Place

CorePNG v0.8.2

CorePNG is a lossless codec based on PNG. Essentially, each frame is compressed as a PNG, so if PNG does it, this codec does too. (RGBA) This codec also has the ability to write P frames and to autodetect when it should. The P frame takes the difference of the previous frame and the current frame and encodes that as a PNG.

CorePNG was originally developed for use with the subtitles, it compresses cartoons and CGI very well and in many cases better than Huffyuv and Loco at the expense of time.

Download
CorePNG-VFW-v0.8.2 Installer (with Source Code) (427KB)
png2avi2png (89KB)

Compression Tests
Using a 31.44 second DivX 3 Simpsons clip.
01:47.265 - 02:18.472 S03E17 - 8F13 Homer at the Bat
512x384 pixels
31.44 seconds
752 frames
No audio
RGB24 Comparison
158,184KBHuffyuv
131,190KBLoco
130,582KBAll I frames Fastest
113,192KB All I frames Normal
109,812KB All I frames Best
73,654KB 2 P frames Normal
61,868KB 100 P frames Fastest
50,369KB 100 P frames Normal
49,189KB 100 P frames Best
40,603KB 1000 P frames Best with Auto-Delta frames

Posted by Jory Stone on Friday, November 28. 2003 at 12:15 in CorePNG
Defined tags for this entry: ,

Trackbacks
Trackback specific URI for this entry

No Trackbacks

Comments
Display comments as (Linear | Threaded)

Jason says,

Thursday, August 6. 2009 at 10:41 (Reply)

Does this support P frames in that regions of change are encoded, then compressed? Such that you start with a I frame and then when identifying deltas you are able to identify regions, then only PNG encode those regions?

I have animations, like the Simpsons, and usually there is only one or two regions of activity. I don't want to encode the entire frame (or entire frame delta) but rather only the regions that is Homer's mouth or whatever else is moving. This seems to be a way to minimize the drawing and encoding.
Jory Stone says,

Sunday, August 16. 2009 at 15:45 (Link) (Reply)

The P frames work by taking the current image and subtracting each pixel with values from the previous frame.
This means that areas that have not changed will be a solid color and compress very well with PNG/zlib.

Thanks,
Jory
Anonymous says,

Monday, August 31. 2009 at 17:20 (Reply)

I installed CorePNG, and can't even select a codec on virtualdub now

"An out-of-bounds memory access (access violation) occurred in module 'CorePNG_vfw'...
...while opening video codec with FOURCC "png1" (compchoose.cpp:166)...
...while enumerating video codecs (compchoose.cpp:154)."

I'm guessing this might be because I'm running Win XP 64-bit edition? Just thought you might want to know.
?r?ola says,

Sunday, February 21. 2010 at 20:48 (Link) (Reply)

I really like this codec, especially the extension to YUY2 and YV12. Is the YUY2 format encoded like RGBA?

Just a thought - the compression could be a lot better if P-frames supported some kind of block-based motion prediction. A heuristic for finding the best vectors would slow it down only a few percent. The motion vectors could be saved as a tiny gray+alpha PNG image.

Add Comment

Enclosing asterisks marks text as bold (*word*), underscore are made via _word_.
Standard emoticons like :-) and ;-) are converted to images.
E-Mail addresses will not be displayed and will only be used for E-Mail notifications