DP Load Image Minimal

The Desert Pixel pack has several image loaders, and this is the bare-bones one: no effects, no batch folder cycling, just load, optionally resize, optionally preserve alpha, and pull whatever metadata is embedded. If you don't need the fancier loaders' extra ports and just want a clean load step, this is the one to reach for.

How it works

image is the standard ComfyUI-input-folder dropdown. resize_image plus width/height gives you an optional resize on the way in, so you're not stuck loading everything at its native resolution if your workflow expects a fixed size. preserve_alpha keeps the alpha channel intact rather than flattening it away - matters if you're loading a PNG with transparency and want to use that transparency as a mask or preserve it through further processing.

Same as the pack's other loaders, it also extracts prompt-style metadata from the source file if it's present - dp_prompt and dp_negative_or_other come back as strings, empty if the file has nothing embedded.

Inputs and outputs

Required:

• image - enum dropdown, ComfyUI's input folder.
• resize_image - boolean, default true.
• width / height - 64 to 2048, default 1024 each, used when resizing.
• preserve_alpha - boolean, default true.

Outputs:

• image - the loaded (and optionally resized) image.
• mask - the alpha-derived mask, if the source has one.
• filename - source filename as a string.
• dp_prompt - extracted prompt text, if present.
• dp_negative_or_other - extracted negative/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 dependencies - a plain image loader.

Common issues & troubleshooting

Transparency isn't coming through. Check preserve_alpha is on (it defaults to true, but if it's been toggled off somewhere) - with it off, the loaded image's alpha channel gets flattened away rather than passed through as a usable mask.

dp_prompt/dp_negative_or_other are blank. These only populate for source files that actually carry embedded generation metadata. A photo, a web-downloaded image, or a re-compressed JPEG (which typically strips that kind of metadata) will just give you empty strings - that's expected, not a fault.

Resize distorts the image. resize_image is a hard resize to width×height, not aspect-preserving - if your source's proportions don't match the target, it'll stretch. Turn resize_image off if you want the image at its native size and aspect ratio untouched.

Wrong loader for the job. If you need a folder of images rather than one file, or you want effects applied at load time, or you want image-format info as an extra output, check the pack's other loaders - DP Load Image Folder, DP Load Image Effects (or its Small variant), and DP Load Image V2 respectively cover those. This one is deliberately the stripped-down option.