colab

input_folder = "/content/drive/MyDrive/AI/StableDiffusion/2023-01/trippy"#@param {type: "string"}
desired_fps = ""#@param {type: "string"}


import cv2
import os

# Get the name of the input folder without the path.
input_folder_name = os.path.basename(input_folder)

# Create the output video path based on the input folder name and desired FPS.
output_video_path = f"{input_folder}/{input_folder_name}_{desired_fps}fps.mp4"

# Retrieve image files with either '.jpg' or '.png' extension and sort them by filename.
image_files = [os.path.join(input_folder, file) for file in os.listdir(input_folder) if file.endswith(('.jpg', '.png'))]
image_files.sort()  # Sort the image files by filename.

# Initialize a list to store the image frames.
frames = []

# Assuming all images have the same dimensions, get width and height from the first image.
first_img = cv2.imread(image_files[0])
height, width, layers = first_img.shape

fourcc = cv2.VideoWriter_fourcc(*'mp4v')  # Change 'mp4v' to the desired codec.

# Convert desired_fps to an integer
desired_fps = int(desired_fps)

video_writer = cv2.VideoWriter(output_video_path, fourcc, desired_fps, (width, height))

# Loop through the sorted image files, read each image, and add it to the frames list.
for image_file in image_files:
    img = cv2.imread(image_file)
    frames.append(img)

# Write the frames to the video.
for frame in frames:
    video_writer.write(frame)

video_writer.release()