Marin85
21st March 2012, 02:11
Hi,
I have started reading up a bit on the topic about deinterlacing, blending and (inverse) telecine, it appears a bit overwhelming at the moment, but I have decided to practice with some samples from my DVD collection.
I have created a sample of a video I have where I am not sure what exactly is going on, so some help would be much appreciated. You can download the sample from here (http://ifile.it/h3qubfv).
It is a 25 fps PAL sample that is reported by mediainfo, DGdecode and Avisynth Info() as interlaced and top field first. However, when I did the AssumeTFF().SeparateFields() and AssumeBFF.SeparateFields() tests, they both seemed inconclusive in my eyes, meaning no obvious discrepancies in the motion except for the "doubling" of each frame (odd and even lines, respectively, due to SeparateFields(), I assume). Does it mean that the clip is actually fake-interlaced? If not, what is it then? And how can it be deinterlaced properly? Should I just feed it to QTGMC() and forget about all that? :o
I have started reading up a bit on the topic about deinterlacing, blending and (inverse) telecine, it appears a bit overwhelming at the moment, but I have decided to practice with some samples from my DVD collection.
I have created a sample of a video I have where I am not sure what exactly is going on, so some help would be much appreciated. You can download the sample from here (http://ifile.it/h3qubfv).
It is a 25 fps PAL sample that is reported by mediainfo, DGdecode and Avisynth Info() as interlaced and top field first. However, when I did the AssumeTFF().SeparateFields() and AssumeBFF.SeparateFields() tests, they both seemed inconclusive in my eyes, meaning no obvious discrepancies in the motion except for the "doubling" of each frame (odd and even lines, respectively, due to SeparateFields(), I assume). Does it mean that the clip is actually fake-interlaced? If not, what is it then? And how can it be deinterlaced properly? Should I just feed it to QTGMC() and forget about all that? :o