ComfyUI Node: Image Loader Crawl (CRT)

Image Loader Crawl (CRT):

The ImageLoaderCrawl node is designed to efficiently load and process images from a specified directory, offering a smart and streamlined approach to image handling. This node is particularly beneficial for AI artists who need to work with large collections of images, as it can automatically scan directories, including subfolders if desired, to identify and load image files with supported formats such as .png, .jpg, .jpeg, .bmp, .tif, .ti, .gi, and .webp. The node ensures that images are converted to a consistent RGB format, which is crucial for maintaining color accuracy across different image types. Additionally, it incorporates a caching mechanism to improve performance by storing file lists and modification times, thus avoiding unnecessary rescans of directories unless changes are detected. This feature not only speeds up the image loading process but also provides a self-healing capability by handling missing files gracefully, ensuring that your workflow remains uninterrupted.

Image Loader Crawl (CRT) Input Parameters:

folder_path

The folder_path parameter specifies the directory from which images will be loaded. It is crucial for defining the source location of your image files. This parameter directly impacts the node's ability to locate and process images, as it determines the starting point for the directory scan. There are no specific minimum or maximum values, but it should be a valid directory path.

seed

The seed parameter is used to select a specific image from the list of available images in the directory. By using a seed value, you can ensure consistent and repeatable selection of images, which is particularly useful for experiments or demonstrations where the same image needs to be used repeatedly. The seed value should be an integer, and while there is no strict minimum or maximum, it should be within the range of the number of images available.

crawl_subfolders

The crawl_subfolders parameter is a boolean option that determines whether the node should include subdirectories in its search for images. When set to True, the node will recursively scan all subfolders within the specified folder_path, allowing for a more comprehensive image collection. This is particularly useful when your images are organized in a nested folder structure. The default value is typically False.

remove_extension

The remove_extension parameter is a boolean option that specifies whether the file extension should be removed from the image's base name when it is returned. This can be useful for simplifying file names or when the extension is not needed for further processing. The default value is usually False.

Image Loader Crawl (CRT) Output Parameters:

img_tensor

The img_tensor output is a tensor representation of the loaded image, formatted as a PyTorch tensor. This format is essential for further processing in machine learning models, as it allows for efficient manipulation and computation. The tensor is normalized to have values between 0 and 1, ensuring compatibility with most neural network architectures.

base_name

The base_name output provides the name of the image file, either with or without the extension, depending on the remove_extension parameter. This is useful for tracking and identifying images during processing, especially when working with large datasets.

num_files

The num_files output indicates the total number of valid image files found in the specified directory. This information is valuable for understanding the scope of the dataset and for debugging purposes, as it confirms the number of images available for processing.

Image Loader Crawl (CRT) Usage Tips:

• Ensure that the folder_path is correctly set to the directory containing your images to avoid errors in loading.
• Use the seed parameter to consistently select the same image for repeatable experiments or demonstrations.
• Set crawl_subfolders to True if your images are organized in a nested folder structure to ensure all images are included.
• Consider setting remove_extension to True if you need simplified file names for further processing or display.

Image Loader Crawl (CRT) Common Errors and Solutions:

Error: File '<selected_file>' was in cache but not found on disk.

• Explanation: This error occurs when a file that was previously cached is no longer available on the disk, possibly due to deletion or movement.
• Solution: Ensure that the file exists in the specified directory. If the file was intentionally removed, clear the cache or rescan the directory to update the cache.

Error loading image '<selected_file>': <error_message>

• Explanation: This error indicates that there was an issue loading a specific image file, which could be due to file corruption or unsupported format.
• Solution: Verify that the image file is not corrupted and is in a supported format. If the issue persists, try converting the image to a different format and reattempt loading.