Converting Video

[PNA]

I have several videos in the .AVI format taken with Canon PowerShot A60 and would love to pull out individual shots for processing.

Anyone with suggestion, please???? :er:

 

[PNA]

I found a way through Windows Movie Maker, but the resolution is poor.

Any others?

 

[markc]

Resolution is going to be bad taken from a video. It's going to be no better than the video's resolution, and there will compression artifacts and other things involved, like interlacing. Vid caps generally suck unless you are dealing with Hi-Def, and there isn't much of that around in the consumer market, if any.

 

[PNA]

Ah, well.....so much for reproducing decent images from video.

Thanks.

 

[guitarkid]

i produce good images from video all the time. avi has no compression but i'm also shooting with canon GL1 3 chip cams. i capture still frames from video while they are on the timeline in my editing software. as far as cheap software to do this, i'm not sure. also if you took video off that cam, it won't be the best.

 

[markc]

AVI files use a bunch of different codecs. You can't go by the file extension. A lot use compression. And "good" really does depend on who is doing the judging and what you want to use it for.

 

[guitarkid]

the avi's i work with have no compression. you're probably thinking of bit rate or frame rate.

 

[markc]

Nope, codec, like Cinepak and MPG-4. "Codec" stands for compression/decompression.
http://www.jmcgowan.com/avicodecs.html

Yours are probably compressed. The GL1 cams are mini DV.

"DV isn't just a tape format; it's also a compression method. All DV camcorders and decks record to tape using DV compression. When you capture DV over Firewire, you're not actually digitizing the video but rather simply performing a file transfer of the DV compressed files. Since everything on a DV tape is compressed using DV compression, there is actually no such thing as "uncompressed DV," yet the term uncompressed DV is often casually used to refer to video that has not undergone any further compression besides its native DV compression."

http://people.csail.mit.edu/tbuehler/video/codecs/avi.html

As far as a source goes, you've got a very good one, but it's still video, which will be interlaced, and your max size is 720x480 pixels. I don't see how that can compare to a real digital still camera or scanned film.
http://www.bealecorner.com/gl1/res/gl1res.html
Like I said, it depends on what you want to use it for. It may be fine for some web images, but not for printing; at least not anywhere near close to what I want for printing, and that's from an very good video camera. Even if you had a super-expensive HDTV camera, that's at 1920x1080. That's starting to get into the range where you could make 4x6 prints.

 

[guitarkid]

i never said it compares with digital photos from a DSLR. i just stated that the avi's i work with utilize no compression and i get great results for what i need them for in video production and web stuff. there is no way any captured still will compare with a dSLR

 

[JoeVanCleave]

... i just stated that the avi's i work with utilize no compression ...

As long as it works for you, that's what matters.

But just for the record, the mini-DV format is based on MPEG-2, which has 12 different compression levels, based on the hardware and software configuration chosen. These various compression levels of MPEG-2 range from sub-VHS resolution all the way up to 720p or 1080i.

It's important to be aware of these differences, for instance when considering purchase of a supposed "HD" television. The MPEG-2 decoder chipset in those TV's determines what's the highest resolution format that it will handle, and the local broadcaster determines at what resolution the signal actually gets broadcast at. I hear people say all the time that "the game's being broadcast in HD", when in fact it's not "HD" (1080p), but merely a lower resolution along the sliding scale that is MPEG-2 compression.

As for mini-DV, it's compressed at 4:1:1; DVC-PRO is 4:2:2, and some of the "broadcast" video formats are 4:4:4. What these numbers mean: the first number tells how many times the luminance (B/W) part of the signal is sampled; the second and third numbers tell how many times the chroma (color) parts of the signal are sampled. So with mini-DV, the color information is sample only 1/4 as much as with a broadcast-qualilty format using 4:4:4. And that's one reason why mini-DV is "compressed". MPEG-2 compression also doesn't transmit every video frame in its entirety, just the time and motion-based differences between full frames. The degree to which it transmits full frames is part of the determination of its video quality. In the lowest standards, a full frame is broadcast once every 4 frames or less, with higher resolution standards sampling full frames almost every frame.

It's important to note that video editing software can only perform edits on full frames, not partial (i.e. "i" and "p" frames [iterative and predictive}).

The higher the sample rate, the more bandwidth taken up, so it's always a balance between bandwidth and quality. For tape-based formats like mini-DV, these standards are set so that the signal will fit within the hardware parameters of the tape format itself.

 

[guitarkid]

thanks for pointing out all the technical jargon. you don't have to impress anyone here if that is what you're going for. i just simply said that a frame capture in editing software is good enough for simple photo files. i'm aware of all the other redtape involved with video as i have been involved in it for about 15 years now. thanks again though.
ciao