Passing batch of images for faster processing about mtcnn

[ipazc]

Hello @Neltherion . Right now there isn't a way to process a batch of images, but I'll work on it.
Thank you for the suggestion!

[chaudharyachint08]

@Vermeille something like 40-60ms on 1920x1080

I tried on GPU instance on Google-colab. For 1280x720 image it took around a second, not at all fast as I was expecting initially

%time _ = face_detector.detect_faces(image)

CPU times: user 1.11 s, sys: 56.2 ms, total: 1.17 s
Wall time: 939 ms

Is it ok to have this high number??

[chaudharyachint08]

@ipazc Still waiting for Batch level implementation, see if you can help

[Neltherion]

@ipazc Thanks. I'll be watching your progress...

[Vermeille]

I'm curious, how fast is it on GPU, without batches?

[YaroslavSchubert]

@Vermeille something like 40-60ms on 1920x1080

[SuperBo]

I also want this features, when will batch of images support be available?

[hbellafkir]

@ipazc are you still woking on it, or its a dead thing ?

[Vinayrraj]

Hello @Neltherion . Right now there isn't a way to process a batch of images, but I'll work on it.
Thank you for the suggestion!

Hi @ipazc, if possible will you be able to suggest by when we can expect this feature?