[英]How to speed up object detection while using faster rcnn/ ssd models
I've got 50 videos spanning 90 minutes each.我有 50 个视频,每个视频跨越 90 分钟。 Now I'm running SSD for object detection and saving the frames with objects and their timestamps in a csv file.
现在我正在运行 SSD 进行对象检测,并将包含对象及其时间戳的帧保存在 csv 文件中。 But the problem is that it still takes entire 90 minutes for each of the videos.
但问题是每个视频仍然需要整整 90 分钟。 My question :- Is there any way by which I can somehow kind of fast forward the SSD object detection model so that each video gets done in let's say 10/20 minutes maybe ?
我的问题:- 有什么方法可以让我以某种方式快速推进 SSD 对象检测模型,以便每个视频都可以在 10/20 分钟内完成?
Is every single frame really matter to you?每一帧对你来说真的很重要吗? If not, just run the detection every x frames only, not on every frame.
如果没有,只需每 x 帧运行一次检测,而不是每帧运行一次。 for example use:
例如使用:
if (frame%2 == 0)
{
run detection
}
for taking only every two frames (even number frame)只拍摄每两帧(偶数帧)
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.