git-hamza

video_1_path = "1.mp4"
video_2_path = "2.mp4"

paths = [video_1_path, video_2_path]

vid_caps = {idx+1: cv2.VideoCapture(path_) for idx, path_ in enumerate(paths)}

tracebar_video = 1
tracebar_frame = 15000

git-hamza

"""
Annotate a video in Cloud Storage provides you an option to generate the output as json.

This script can be used you annotate a local video (https://cloud.google.com/video-intelligence/docs/logo-recognition#annotate_a_local_video).
This script will help to generate the bounding box per frame and also draw it on the video.
"""

import io
import json
from google.cloud import videointelligence

git-hamza

import yaml
import os
import matplotlib.pyplot as plt

# Define paths to image and label directories, and class YAML file
data_dirs = ["train", "valid", "test"]
class_yaml_path = "data.yaml"

# Load class names from YAML file
with open(class_yaml_path, 'r') as f:

git-hamza

SSH to the remote server:

Open cmd/terminal and connect the local port to the remote server port. ssh -L 127.0.0.1:3000:0.0.0.0:1234 USER@IP

Start the docker container inside the remote server:

For starting the docker, make sure to keep the port same as provided above. An example of starting a docker is given below. docker run -it -p 1234:8888 IMAGENAME /bin/bash

Run Jupyter Notebook:

Now that you have run the above command with /bin/bash, it will take you inside the docker container. Make sure to have jupyter notebook installed in there. Run the following command inside the docker container terminal.