Video Editor Utility
Source code for the utilities is available here
The setup process involves building a docker image from inside the source code folder like below.
docker build --tag platerecognizer/video-editor .
The utilities work using a config.ini
that is in a similar format to the one used by Stream.
#
Extract Frames as ImagesAdd [[[frames]]]
under a camera section
[cameras] [[camera-1]] active = yes url = /user-data/test.mp4 # Extracting Frames [[[frames]]]
The extracted frames will be saved in a folder named {camera_id}_frames in the folder containing your config.ini.
#
Draw Vehicle and Plate Bounding BoxesThis works by uploading every frame in the video to Snapshot.
Visualize plates and vehicles on a video, add [[[visualization]]]
under a camera section
[cameras] [[camera-1]] active = yes url = /user-data/test.mp4 # Visualization [[[visualization]]] token = 'ddjjdkdkfjjf22333############' sdk_url = 'http://localhost:8080'
A new video named {camera_id}_visualization.avi will be created in your Stream folder.
Token is optional if using the Snapshot SDK.
#
Blur Plates and FacesThis works by uploading every frame in the video to the Blur.
To blur plates and faces in a video, add [[[blur]]]
under a camera section
[cameras] [[camera-1]] active = yes url = /user-data/test.mp4 # Blur [[[blur]]] blur_url = 'http://localhost:8080'
A new video named {camera_id}_blur.avi will be created in your Stream folder.