olafenwamoses / traffic-net Goto Github PK

imageai ==2.0.3
tensorflow ==1.15.0
h5py ==2.10.0
opencv-python
keras

Hi @OlafenwaMoses, your project is great. I found your video demo in real-time is really amazing, but the code you have posted right now just processes the image. Could you share the code that I can make the same thing like you did in your video demo, thank you.

Test the new picture you gave is not accurate，Not like the result you gave

I get wrong results when i test the model with video file, but not, if i give pictures to the model. can you please show me, how can i make prediction for video frames.
results look like:
Fire : 100.0
Sparse_Traffic : 0.0
Dense_Traffic : 0.0
Accident : 0.0
video file is: https://www.pexels.com/video/traffic-flow-in-the-highway-2103099/

my Code:

Hi. I'm trying to run this system on tensorflow-gpu, but it raises some exceptions with cudnn (tf-gpu 1.14; CUDA 10.0; cudnn 7.4.2). When i run it on CPU, instead, no problem appears. So, it is possible to run it on GPU or not?

Hello Olafenwa!! I ran the pre-trained model and that works really great on images. Can you also specify that how can we test the pre-trained model on a video file?