IVTC of unknown framerate using unknown cadence?

[knutinh]

Hello

I am trying to get the full frames of a set of nominally 18 (?) fps 8mm converted to 50 fields per second DV. I have inspected the output frame-by-frame in virtualdub (is that field(N) + field(N+1) waved into one frame?), and could not find some simple pattern of combed/non-combed frames.

I dont know the equipment used, or what cadence was used. Is there any tool that will do this robustly and automatically? I thought about writing one myself, but I figured that someone surely must have thought it more through than me:
1. input approximate source framerate, N/D
2. input "field search window", M
3. input difference tolerance, e

Then the algo would look M odd/even fields before or after each field to find fields that are equal (within tolerance e). Any duplicated even or odd fields should then be easily (and inexpensively) found?

In my case, I was hoping for a sort-of bulletproof way of getting back my 18fps progressive film from the 50i container.

-k

[knutinh]

Thank you. Please see the link for ~100 frames of video.

http://www.sendspace.com/file/a7cnp9

[knutinh]

Great. Thanks. Do you think that there is any benefit from averaging duplicated fields instead of dropping them to reduce compression artifacts?

http://www.adamwilt.com/DVvsMJPEG.html

The following discussion by Guy Bonneau was developed from a posting on DV-L in August of 1998.
...
Besides, the DV standard requests that the video stream be compressed on a frame basis, while most MJPEG codecs compress the video stream on a field basis. Because of the interlaced nature of a frame, the two merged fields may contain quite different information if the scene has any motion, and this gives rise to a problem when the pixel data are processed by DCT. In this case, the standard DCT algorithm used with MJPEG will perform badly because of the mixing of different field information. The standard DCT will produce a lot of inflated AC coefficients (related to the spurious high-frequency vertical detail caused by the differing interleaved fields) which are very difficult to compress efficiently even with high quantization values.

To overcome this problem the DV standard make a provision for a new DCT "mode" which is fine tuned for this situation and doesn't produce inflated AC coefficients. Rather than process the pixel data on a 8 x 8 block basis with the standard DCT, the new DCT mode deinterlaces the pixel data in two independent 4 X 8 blocks and
computes two DCTs on the 4 X 8 pixel blocks, still producing a total of 64 DC and AC coefficients. The DV jargon defines this new DCT mode the "2-4-8-DCT mode" versus the standard "8-8-DCT mode".

[Fyx]

I had decent luck with the following for 25>18 on your clip:


directshowsource("D:\seg.avi")
converttoyv12(interlaced=true)

tfm(mode=0, slow=2) #deinterlace by field matching
tdecimate(cycleR=7, cycle=25) #remove 7 most similar frames every 25


Filters are from Tritical's TIVTC here: http://web.missouri.edu/~kes25c/

[knutinh]

There were some posts in this thread that disappeared. What happened?

-k

[Wilbert]

It looks like Stephen R. Savage deleted most of his posts. Was there something interesting in them (then i can put them back)? I can't send him a PM, so i can't ask him about his reasons.