Does 4.8.0 not support AVIF imencode/decode?

Python
1

import cv2
import socket
import numpy as np

print(cv2.version)

Open the webcam

cap = cv2.VideoCapture(0) # Change the index if you have multiple cameras

while True:
# Capture a frame from the webcam
ret, frame = cap.read()

# Encode the frame as jpeg, tiff, webp image
# Set the quality (optional)
encode_param_jpeg = [int(cv2.IMWRITE_JPEG_QUALITY), 50]
result_jpeg, encimg_jpeg = cv2.imencode('.jpg', frame, encode_param_jpeg)

#tiff
encode_param_tiff = [int(cv2.IMWRITE_TIFF_COMPRESSION), 50]
result_tiff, encimg_tiff = cv2.imencode('.tif', frame)

#webp
encode_param_webp = [int(cv2.IMWRITE_WEBP_QUALITY), 50]
result_webp, encimg_webp = cv2.imencode('.webp', frame, encode_param_webp)
print("frame size: ", frame.shape)

# #avif
# encode_param_avif = [int(cv2.IMWRITE_AVIF_QUALITY), 50]
# result_avif, encimg_avif = cv2.imencode('.avif', frame, encode_param_avif)



#comparethe size of the frame and encoded image

print("jpeg size: ", len(encimg_jpeg))
print("tiff size: ", len(encimg_tiff))
print("webp size: ", len(encimg_webp), "\n")

# decode the jpeg encoded image
decimg_jpeg = cv2.imdecode(encimg_jpeg, 1)
decimg_tiff = cv2.imdecode(encimg_tiff, 1)
decimg_webp = cv2.imdecode(encimg_webp, 1)

# merge the images   side by side
merged_image = np.hstack((decimg_jpeg, decimg_tiff, decimg_webp))






# Display the frame
cv2.imshow('frame', merged_image)
cv2.waitKey(1)  # Adjust as needed to display the frame

I am trying to reduce the amount of data in video transmission.

If you try to compress the image with the AVIF codec by activating lines 29-30 in the above code, the following error occurs.

result_avif, encimg_avif = cv2.imencode('.avif', frame, encode_param_avif)

cv2.error: OpenCV(4.8.0) /io/opencv/modules/imgcodecs/src/loadsave.cpp:1120: error: (-2:Unspecified error) could not find encoder for the specified extension in function ‘imencode’

I am using opencv-4.8.0 in a python3 environment.
I’m asking because I think there is a variable for the avif codec in the documentation.
https://docs.opencv.org/4.8.0/d8/d6a/group__imgcodecs__flags.html

2

that is not done by compressing pictures. that is done by video compression.

AVIF support is optional and requires the right libraries to be available when you build your OpenCV.

3

thanks for the information. I was using opencv-python package.
maybe I have to build it from source like old time.

4

AVIF support is extremely new. you’ll probably need libavif. I don’t know if the cmake stuff fetches that or requires you to provide it.

also looks like you’ll have to use VideoWriter, since that format is based on AV1, a video codec.