DP Load Image Effects

Instead of loading an image and then chaining four separate effect nodes off it, this one folds the load-and-fork pattern into a single node: pick an image, choose four different effects, get four differently-processed versions back in one step. It also pulls whatever generation metadata is embedded in the source file, which is a nice bonus if you're working from a previously-generated image rather than a plain photo.

How it works

image picks a file from ComfyUI's input folder the normal way. From there, four independent effect slots (effect_A through effect_D) each apply one pick from the same 24-option effect menu the pack's effect-processor nodes use - grayscale, enhance, flips, rotations, posterize, sharpen, contrast, equalize, and roughly a dozen more not individually named in the node's brief. A shared effect_strength scales all four. Optionally, resize_image (on by default here) forces all four outputs to a common width/height, which keeps your four variants consistent in size if you're about to grid or batch them.

The metadata pull is the other half of what this node does: filename, dp_prompt, and dp_negative_or_other come out as strings alongside the four images. If your source image was generated by ComfyUI or a similar tool that embeds its generation parameters, this reads that data back out - useful for reusing or referencing the prompt that made a reference image you liked, without having to dig through your own generation history to find it.

Inputs and outputs

Required:

• image - enum dropdown, populated from ComfyUI's input folder.
• effect_strength - 0 to 1, default 1.
• effect_A through effect_D - each independently pick from 24 effects, default original (no-op).
• resize_image - boolean, default true.
• width / height - 64 to 2048, default 1024 each, used when resize_image is on.

Outputs:

• image_A through image_D - the four processed variants.
• filename - the source filename as a string.
• dp_prompt - extracted prompt text, if present in the image's metadata.
• dp_negative_or_other - extracted negative prompt (or other) text, if present.

How to install it

Search ComfyUI-Desert-Pixel-Nodes in ComfyUI Manager, or:

cd ComfyUI/custom_nodes
git clone https://github.com/DesertPixelAi/ComfyUI-Desert-Pixel-Nodes

Restart, find it under DP/image. No models, no extra dependencies - the "load an image from a folder" mechanic reuses ComfyUI's standard input-folder handling, and the effects are pure image math.

Common issues & troubleshooting

dp_prompt and dp_negative_or_other come back empty. These only populate if the source image actually has generation metadata embedded - a plain photo, a screenshot, or a JPEG with metadata stripped (JPEG re-compression and many editing tools drop PNG metadata entirely) will just give you empty strings. That's expected, not a bug: there's nothing to extract if the file never had it.

One of the four output slots looks unchanged from the source. Check that slot's effect letter isn't left on original - that's the default for all four, and it's a genuine passthrough. It's easy to set A, B, and C deliberately and forget D was never touched.

Resize is squashing your image instead of fitting it. resize_image is on by default here (unlike the pack's smaller effect processor, where it's off by default) and forces a hard resize to width/height - not an aspect-preserving fit. If your source image's aspect ratio doesn't match your target dimensions, expect stretching. Turn resize_image off if you want to keep the source's native size and aspect ratio.

Dropdown doesn't show a file you just added. The image dropdown is populated from ComfyUI's input folder at the time the UI loads it - files added to the folder after that may need a page refresh or an upload through ComfyUI's own file picker to show up, same as any standard image-loading node.