Save 4 hours! We auto-setup your workflow! Free!

Drop your workflow.json — we handle every dependency, custom node, and model. Just open the link and run.

Auto-Setup Workflow Json (Free) Now!
ComfyUI > Nodes > FM_nodes > ConvIR Motion DeBlur

ComfyUI Node: ConvIR Motion DeBlur

Class Name

ConvIR_MotionDeBlur

Category
FM_nodes/ConvIR
Author
fuoum (Account age: 2555days)
Extension
FM_nodes
Latest Updated
2025-03-27
Github Stars
0.01K

How to Install FM_nodes

Install this extension via the ComfyUI Manager by searching for FM_nodes
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter FM_nodes in the search bar
After installation, click the Restart button to restart ComfyUI. Then, manually refresh your browser to clear the cache and access the updated list of nodes.

Visit ComfyUI Online for ready-to-use ComfyUI environment

  • Free trial available
  • 16GB VRAM to 80GB VRAM GPU machines
  • 400+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 200+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

ConvIR Motion DeBlur Description

Reduces motion blur in images or batches using a selected ConvIR model to improve clarity.

ConvIR Motion DeBlur:

ConvIR_MotionDeBlur uses a ConvIR image-restoration model to reduce motion blur. It is intended for images where movement during capture has softened edges or smeared detail. The node runs the selected deblurring model on each image in a batch and returns the processed images. Results depend on the selected model and the source material: it may improve visible clarity, but it cannot guarantee recovery of detail that is absent from a heavily blurred image.

ConvIR Motion DeBlur Input Parameters:

src_img

The IMAGE to process. You can supply a single image or a batch; the node processes each batch item separately and keeps their order. This input determines the content and dimensions of the result. No minimum, maximum, or default image size is specified in the provided node interface. Larger images or batches may require more processing time and memory.

model_name

Selects the deblurring model used for processing. The node looks up the corresponding model file and loads it before processing the images, so your choice can affect the appearance of the result. This is a dropdown whose options come from DEBLUR_MODELS and whose default is DEBLUR_DEFAULT. The provided context does not include the values of those constants, so the exact option names and default label cannot be listed here. There is no numeric minimum or maximum.

ConvIR Motion DeBlur Output Parameters:

res_img

The processed IMAGE. It contains one result for each input image, in the same batch order. The node crops the model’s prediction to the original image height and width, then limits its values to the range 0–1 before returning it. You can compare this output with the original to judge whether blur has been reduced without introducing unwanted changes.

ConvIR Motion DeBlur Usage Tips:

  • Start with the default model_name, then compare the result with the original at the same zoom level, especially around edges and fine textures.
  • If your installation offers several model_name choices, try them on the same image and keep the result that best balances clarity with a natural appearance. The provided source does not describe what distinguishes those choices.
  • For a batch, check more than the first result. The node processes images individually, and blur severity may differ across the batch.
  • If processing runs out of memory, try a smaller image or a smaller batch. The node uses CUDA when available and otherwise runs on the CPU; the interface does not expose a device or model-size setting.

ConvIR Motion DeBlur Common Errors and Solutions:

FileNotFoundError for a model checkpoint

  • Explanation: The node builds a path to the selected model file and loads it. This error can occur if that file is missing or the installation’s model directory is incorrect.
  • Solution: Check that the checkpoint required by the selected model_name is installed in the expected model directory. Reinstall or restore the model file if necessary.

KeyError: 'model'

  • Explanation: The node expects the loaded checkpoint to contain an entry named model. A different checkpoint format may not contain it.
  • Solution: Use the checkpoint supplied for this ConvIR node and model choice, rather than a file from another model or an incompatible release.

RuntimeError: Error(s) in loading state_dict for ConvIR

  • Explanation: The checkpoint’s saved model weights do not match the ConvIR model the node creates. This can happen when the wrong checkpoint is installed for a model choice.
  • Solution: Confirm that the selected model_name resolves to its intended checkpoint. Replace any mismatched file with the correct one.

RuntimeError: CUDA out of memory

  • Explanation: The image, batch, and model exceed available GPU memory during processing.
  • Solution: Reduce image dimensions or process fewer images at once, then try again. Close other GPU-intensive tasks if practical.

Error while reading or loading the checkpoint

  • Explanation: torch.load may fail if the model file is incomplete, damaged, or incompatible with the installed PyTorch environment. The exact error text depends on the cause.
  • Solution: Restore the correct checkpoint from a trusted source and check that your installation meets the node’s dependency requirements.

Error during image preprocessing or model processing

  • Explanation: The node expects a compatible ComfyUI IMAGE input. An incompatible tensor, unexpected image data, or a processing failure can stop execution; the exact message depends on where it fails.
  • Solution: Connect a standard IMAGE output, test with one ordinary image, and check the full error trace to identify whether preprocessing or model processing failed.

ConvIR Motion DeBlur Related Nodes

Go back to the extension to check out more related nodes.
FM_nodes
RunComfy
Copyright 2025 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.

ConvIR Motion DeBlur