Michael Welter

Nevermind! My branch was outdated. I think your last patch with the scale factor fixed most of my problems.

P.S. In my own auto-labeling script I was using the MTCNN detector because it is much faster than FAN and consumes less memory. But I also got the failures I...

I use the pretrained weights, too. As I wrote, using the most recent master branch worked for me. The DECA network does not like the way MTCNN picks the bounding...

There is my fix attempt https://github.com/cnr-isti-vclab/vcglib/pull/24

Hi, tbh I don't know. But I guess not. It seems there are a few more steps required as I found in this blog https://habr.com/en/company/intel/blog/575632/. It is also not planned....

Hi I haven't published an example file. The name is just a placeholder. You'd have to capture one by yourself with raspivid. Or maybe you can extract the raw h264...

I'm not surprised at all as I haven't seen many people with glasses on in the datasets. For now it's an unfortunate limitation. But thank you for confirming. I'll keep...

@sthalik I labeled the ticket as `feature` (request?). I guess it would be open for a while ... maybe useful to collect reports. But you could close it as well...

Maybe this wasn't the issue. I also get a crash in the code that get the camera names. Not sure what is wrong. Perhaps a broken build?

Win32. And there is another thing I'm wondering about. You are still linking to the static runtime DLLs, right? So what if some object changes ownership across DLL boundaries, won't...