Tensorflow Implementation of YOLO v1 Object Detection. Paper: arxiv.org/abs/1506.02640 [C] Project: github.com/pjreddie/darknet [Tensorflow] Project: github.com/dshahrokhian/YOLO_... Song: Voicians - Survival League
Пікірлер: 17
@levlev.10283 жыл бұрын
not sure how this ended up in my yt recommendations but it seems cool asf 😅
@FilterYT7 жыл бұрын
This is very cool, thanks for sharing!
@dshahrokhian7 жыл бұрын
Thanks for watching.
@Zlx-sx9gt2 жыл бұрын
Amazing 👌👌👌👌
@lujia6 жыл бұрын
Is there a way to ease the sudden change in bounding box size?
@larryguo25297 жыл бұрын
Tks for posting, great sharing. I got some info that without a GPU, it may take 1-2s per frame, What If I only like to identify certain categories ? (say, only car, and motocycles), could it increase the speed (without GPU) ?
@dshahrokhian7 жыл бұрын
The inference will take the same time since the only difference is the number of categories that you trained the network with. The architecture of the network remains the same, thus the elapsed time from input to output will also remain the same.
@larryguo25297 жыл бұрын
ok, tks for the reply.....(still thinking a way to increase the speed on cost economic embedded platform )...
@jochem33686 жыл бұрын
Is it possible to assign two categories to 1 objects for example: car and yellow
@seenPunktSee2 жыл бұрын
spot the *horse*
@FastNotSave7 жыл бұрын
So this network is labelling a Transformer as a Person. Interesting :D
@dshahrokhian7 жыл бұрын
Or a motorbike (0:30).
@willysatrionugroho80864 жыл бұрын
That's because transformers so ambiguos like a shemale. Shemale have 50% chance as a male or a female unless you see his adam apple.