Video streaming in Tkinter with Python

Hey programmer, This tutorial will help you with the concept of video streaming in Tkinter. We will achieve this by using a Frame and a Label widget and use some libraries of Python namely: Tkinter, PIL, and Imageio.

Using:

from tkinter import *
import imageio
from PIL import Image, ImageTk

Video streaming inside a frame in Tkinter

We will use the Imageio library to store a video and get its metadata and also its frames. Then we will use the PIL library to add that frame captured as an image in the Label() widget defined inside a Frame() widget.

The Stream() function:

def stream():
    try:
        image = video.get_next_data()
        frame_image = Image.fromarray(image)
        frame_image=ImageTk.PhotoImage(frame_image)
        l1.config(image=frame_image)
        l1.image = frame_image
        l1.after(delay, lambda: stream())
    except:
        video.close()
        return

The above function pulls each frame out of the video as an image then stores it in a variable and sets the image parameter in a Label. The Stream() function is recursively called. This is done using the .after method()  to register callbacks.

Follow the steps below:

1. Define the Tk window as the root variable.
2. Next, we create a Frame and a Label with the frame as its parent, the latter will be used later to display each frame of the video as an image.
3. Then we define a variable to store the path of the video and call the .get_reader() method of imageio library with the previous variable as a parameter to read that video.
4. Next, We calculate the FPS of the video by .get_meta_data()[‘fps’] method and do some calculations to the result for accuracy.
5. In the next step, we call the stream() function, which streams the video.
6. End the program with a root.mainloop() to keep the process going recursively as long as the window runs.

The whole program looks like so:

from tkinter import *
import imageio
from PIL import Image, ImageTk
def stream():
    try:
        image = video.get_next_data()
        frame_image = Image.fromarray(image)
        frame_image=ImageTk.PhotoImage(frame_image)
        l1.config(image=frame_image)
        l1.image = frame_image
        l1.after(delay, lambda: stream())
    except:
        video.close()
        return   
########### Main Program ############

root = Tk()
root.title('Video in a Frame')
f1=Frame()
l1 = Label(f1)
l1.pack()
f1.pack()
video_name = "Futurama.mkv"   #Image-path
video = imageio.get_reader(video_name)
delay = int(1000 / video.get_meta_data()['fps'])
stream()
root.mainloop()

To learn more about Tkinter:

Introduction to Tkinter module in Python