Video Predictor

Easy way to run models on video inputs and export visualized results as videos.

Usage

from BiWAKO import MODNet
from BiWAKO.api import VideoPredictor


model = MODNet()
video_predictor = VideoPredictor(model)
video_predictor.run("some/video.mp4", title="modnet_prediction.mp4")

BiWAKO.api.VideoPredictor

__init__(self, model) special

Initiaslize VideoPredictor.

Parameters:

Name	Type	Description	Default
model	BaseInference	BiWAKO model instance.	required

Examples:

>>> video_predictor = VideoPredictor(model=BiWAKO.MiDAS())

run(self, video_path, title=None)

Predict video at video_path, render the result as a single mp4 and save the result at the same directory.

Parameters:

Name	Type	Description	Default
video_path	str	Path to the video to predict.	required
title	str	Title of the output mp4 file. Defaults to use the model name.	None

Examples:

>>> video_predictor.run(video_path="video.mp4")