Inference on videos gives black gazemap

[pascalxia]

Inference on videos gives black gazemap

[pascalxia]

Hi Gabriele, Thank you very much for your interest in our work. Is that the case for all the videos of your data or just some of them? Would you mind sharing with me one example video to let me test it on my side? Best, Ye

…

On Fri, Jan 10, 2020 at 9:21 AM barloccia ***@***.***> wrote: Hello, I am trying to implement your work as baseline, but when I run the inference on my own data it gives output black images. I followed the steps described, it seems to be all right. My impression is that there's something missing in the pre-processing phase (so it can be about my data). Could you please give some feedback on it? I really appreciate your work, Bests, Gabriele — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#6?email_source=notifications&email_token=ABV3JLQ3AINA6QECCV3KUI3Q5CVDBA5CNFSM4KFK4SDKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IFM3HNA>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABV3JLV5XAOLF2F3WP47S73Q5CVDBANCNFSM4KFK4SDA> .

[gabarlacchi]

Thanks for the quickly answer!
Unluckily I cannot share the videos for privacy reason, I'll try to find a way in next days if there'll always be need. I hope you understand.

Firstly, the first error message which I receive is the following:
"Lossy conversion from float32 to uint8. Range [0, 1]. Convert image to uint8 prior to saving to suppress this warning."
This happens at the last line of infer.py

Thus I tried to modify the code through the following line:
"gazemap = gazemap.astype(np.uint8)" and then save the image.
What I obtain is a black image.

The extraction of the frames from the video it seems good. I tried with different videos with no differences.
The format is .mp4 (originally was .MOV), but no luck.

[Anguliachao]

Dear Pascalxia @pascalxia ,
I follow the exact instructions written by you in this repo ,but my results turn out to be all black instead of visualization result in your figure(like barloccia's results). Would you mind helping me out ? or maybe reply will it be some other issue in image writing methods. I use imageio instead of scipy(deprecated already), but others remains with the original repo. Many thanks

[fahad-lateef]

Hi Pascalxia,

Really appreciate your work

Same issue with me as @Anguliachao and @barloccia
I have implement your work, followed all the steps mentioned. Everything went good but when run test on my data, I get black images in output.

@pascalxia If you can guide?

[fahad-lateef]

Solved..

[MingiJeong]

@fahad-lateef glad to hear that you solved. I've been struggling with addressing that whole black issue. Could you share a method on how you solve please?