ghosting in old tv clip

[ministrix]

Hello!
Got this sample http://www.megaupload.com/?d=3KHUVBS7 for you guys (meaning girls as well) to chk:) Its a tv clip from '69 so it may be unfixable. The problem is ghosting of her hands when she wavers them. I'm not sure if i should keep it at 29 fps or try 26 fps which is what i guess it was originally shot at. I tried different filters like telecide and field deinterlace (blend=false) but its no good. In some frames it looks like the color of her hand is following her hand when she wavers it around. Any ideas?

[davidhorman]

I'm not sure if i should keep it at 29 fps or try 26 fps which is what i guess it was originally shot at.

I'm not sure what makes you think that - it looks like 29.97fps interlaced video to me.

I don't think there's been any strange conversion - it's just some very old video shot on very old cameras.

David

[ministrix]

I thought it was shot for american tv and therefor thought it was 26 fps but im not sure how they did it back then in us. And do u have any suggestions how to remove the ghostings of the hand? Field deinterlace matches up the frames good but still, its ghosting there and thats what i want to remove.

[neuron2]

American terrestrial NTSC TV is 29.97 fps, not 26 fps.

[ministrix]

OK. Then i need to stick to ~29 fps. But is there a option to remove those artifacts?

[neuron2]

It looks like camera motion blur to me and is hardly noticable when played at normal speed. I would leave it alone.

Perhaps someone else has a different take on it.

[Joel Cairo]

Our original poster has a good eye-- what he's pointing out are (probably needless) DVNR artifacts, that are now burned into the image, and are therefore unfixable. With a lighter corrective touch, or more careful engineering, they'd have been easily avoided.

-Kevin

[2Bdecided]

Are there DVNR processes that are so harsh? It seems strange that the white areas leave a black after-image. That's more reminiscent of certain tube cameras.

Anyway, you want to be glad you have colour video archive footage from 1969 at all. Lots of UK broadcasting was still B&W then, and much of the archive stuff has been destroyed. :( To see a nice clean fluid video from the 1960s is rare over here! Most of what survives is only on 16mm film telerecordings: B&W, flickery, scratchy, one field only or both fields merged, often with horrible optical sound. Still, some people try to rescue it...

http://www.restoration-team.co.uk/

Cheers,
David.

[Joel Cairo]

Unfortunately, the answer to your DVNR question is "Yes". I'm not sure which system it is, but I see that type of artifacting a lot. The embossed horizontal bar on the shot change is another giveaway... it's not a residual image from the vintage video camera switcher-- it's from the DVNR processing.

And as for restoration efforts, thank you for noticing-- there are indeed those of us that are trying to do what we can to improve classic television:

http://www.kinescopes.com/livefeed_reviews.html

:)

-Kevin

[2Bdecided]

Unfortunately, the answer to your DVNR question is "Yes". I'm not sure which system it is, but I see that type of artifacting a lot. The embossed horizontal bar on the shot change is another giveaway... it's not a residual image from the vintage video camera switcher-- it's from the DVNR processing.Obviously I spotted that - that's DVNR?! That's terrible!

And as for restoration efforts, thank you for noticing-- there are indeed those of us that are trying to do what we can to improve classic television:

http://www.kinescopes.com/livefeed_reviews.html
Is that you? I've only seen the download demo, which unfortunately was at 29.97fps progressive 320x240. I'll have to check out the Elvis DVD.

I've seen a lot of the comparable process over here (VidFire) which works well. A poor man's option is to play it through a Philips TV with it's own motion interpolation!

I realise the NTSC source kinescopes are more of (or a different) challenge compared to the PAL telerecordings because you have more fields stored on every second film frame!

Cheers,
David.