Hello, thanks for sharing the great dataset. I am trying to generate my own blur frames using frames in GOPRO_Large_all following the process you described in #59 . However, I met an inconsistency issue of mismatched frame index.
For example, in the GOPRO_Large/test/GOPR0384_11_05, the first blur frame is 004001.png, which if I understand correctly, will be an average of frame GOPRO_Large_all/test/GOPR0384_11_05/044001.png~044011.png. However, when I did the average, the result is not the same, there seems to be a few frames missing.
I also find that GOPRO_Large_all/test/GOPR0384_11_05/044001.png happens to be exactly the same as GOPRO_Large/test/GOPR0384_11_05/sharp/004001.png. Shouldn't the sharp/004001.png be the same as GOPRO_Large_all/test/GOPR0384_11_05/044006.png, since it's the middle of the first 11 frames?
This issue happens also in several other clips. Is there any wrong with my blur generation process?
Hello, thanks for sharing the great dataset. I am trying to generate my own blur frames using frames in GOPRO_Large_all following the process you described in #59 . However, I met an inconsistency issue of mismatched frame index. For example, in the
GOPRO_Large/test/GOPR0384_11_05,
the first blur frame is004001.png,
which if I understand correctly, will be an average of frameGOPRO_Large_all/test/GOPR0384_11_05/044001.png~044011.png.
However, when I did the average, the result is not the same, there seems to be a few frames missing. I also find thatGOPRO_Large_all/test/GOPR0384_11_05/044001.png
happens to be exactly the same asGOPRO_Large/test/GOPR0384_11_05/sharp/004001.png.
Shouldn't thesharp/004001.png
be the same asGOPRO_Large_all/test/GOPR0384_11_05/044006.png,
since it's the middle of the first 11 frames?This issue happens also in several other clips. Is there any wrong with my blur generation process?