search

YuHengsss / YOLOV

This repo is an implementation of PyTorch version YOLOV Series
Apache License 2.0
309 stars 43 forks source link

About the model's random select frame input #17

Closed RachelLYY closed 1 year ago

RachelLYY commented 1 year ago

Hi! I notice that the input of this model is randomly selected from the original video. May I know how you make use of the non-keyframe information? And is the result of this model the full result of the video(including key and non-key frames) or just keyframe results? Thanks!

YuHengsss commented 1 year ago

Hello, all of the frames will act as key frames and they will refer to each other in our implementation and all frames in one batch will have prediction results.

RachelLYY commented 1 year ago

Hello, all of the frames will act as key frames and they will refer to each other in our implementation and all frames in one batch will have prediction results.

Thanks! May I also ask another question: What's the 'Randomly select 𝑓 frames' mean in the model image in the paper?

YilanWang commented 1 year ago

Hello, all of the frames will act as key frames and they will refer to each other in our implementation and all frames in one batch will have prediction results.

Thanks! May I also ask another question: What's the 'Randomly select 𝑓 frames' mean in the model image in the paper?

It is the codes what you see in your question. The author random select f frames in the whole video, and use it to train and test