-
-
Notifications
You must be signed in to change notification settings - Fork 5
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
thread '<unnamed>' panicked at 'called Option::unwrap()
on a None
value', in slideo/crates/matching-opencv/src/lib.rs:371:69 stack backtrace:
#4
Comments
That is interesting. Also, in general, how well does the tool work for you (besides these fixable hickups)? |
Could you please tell me what exactly I should put where? I am not really that familiar with rust and couldn't get it to work with your instructions, sorry. Otherwise the tool works really well. - And it is also only on the lectures I tried to run through slideo I get this specific error (maybe because the person is scrolling backwards in the slides? - I haven't really looked into you logic even though I read the background). Sometimes it also seems to miss some pages if they look to similar to previous ones - at least that's what I think that happens, I don't know for sure. (Example, page 7 of this talk should be found at around minute 7 or 8 but it doesn't (at least for me). -> should I open another Issue for this?) |
Can you reproduce this issue with public slides/videos? This statement instead of line 370 should report what goes wrong: rated_best_matches.sort_by(|a, b| {
b.2.partial_cmp(&a.2)
.expect(&format!("Could not compare {:?} with {:?}", a.2, b.2))
}); I fear further debugging will be required though.
Slideo currently looks at each frame independently - reordering all frames should not change the result. |
If two consecutive frames look too similar (like only 1% of pixels change), the second frame is ignored. This is an optimization right after encoding the video. Does this happen after you removed the sort_by? That sort_by should usually prevent this. Sometimes the transformation is not pixel-perfect and if all candidates are very similar, mismatches can happen. |
Backtrace is the same as before (as it should be), output of println: If you want I could send the files I am trying to process (or a link to them) via email - I don't want to post it here because it is not public. |
Yeah, a link to the files would be helpful! (gh@hediet.de) |
I should add, that I don't get the error running the slideo release in a windows-VM, so the problem seems to be somehow MacOS/Linux-related |
Urgh... I don't have a Mac system available. I can try on WSL, but I don't have a rust system set up there. |
As you stated, I have no problems on windows. I'm positively surprised that slideo can analyze that video! I guess slideo could be optimized specifically for slides that build up incrementally (like slides that are written on video). |
I have found another OS-specific difference: Also I replaced line 370 with the following code (I don't know how much sense this makes, but it seems to work fine and still finding all the slide timestamps even in the videos I sent you):
|
On some Videos/pdf-combinations I get this error on MacOS while slideo is processing frames: (This only happens on certain frames and always on the same ones)
Then, after completely processing the video the webserver isn't started and when the pdf is opened manually by executing
slideo file.pdf
the pdf doesn't have any timestamps.Full Backtrace:
Commenting out line 370 in the mentioined
lib.rs
seems to fix the issue even though I don't know why. The program also seems to still work fine after commenting out this line, because the same thing is written in line 329.The text was updated successfully, but these errors were encountered: