ComfyUI Custom Nodes by Desert Pixel

A collection of custom nodes for ComfyUI focused on animation, image processing, and workflow optimization.

Features

Animation & Video

• Animation Calculator (5 Inputs): Complex animation timing calculator
• Fast/Slow Motion: Control video playback speed
• Video Effects:
  • Sender/Receiver system for effect application
  • Flicker effects
  • Transition effects

Image Processing

• Big Letters: Create text-based images
• Color Analyzer: Analyze and generate color palettes
• Logo Animator: Create animated logos
• Smart Image Saver: Enhanced image saving with metadata
• Image Effects: Various image processing effects including:
  • Sepia
  • Vignette
  • Edge detection
  • Sketch effects
  • Relief shadows
  • And more...

Utility Nodes

• Random Min/Max: Generate random numbers with control
• Random Characters: Generate random characters/strings
• Aspect Ratio Picker: Easy aspect ratio selection
• Clean Prompt: Clean and format prompt text
• Broken Token: Analyze and split Flux prompts
• Quick Model Link: Manage symbolic links for models
• Multi Styler: Apply multiple styles to prompts
• Zero-One Floats: Simple float input/output controls

Full nodes list:

DP_Animation_Calculator_5Inputs

Animation calculator helper for switching between images/styles with img2video & video2video using AnimateDiff. Creates timing data for prompt scheduling, fade masks, and IP Adapter batch processing.

Each image has a keyframe start point and associated prompt text. The node can be used with or without image inputs to generate scheduling prompts or fade mask timing.

Inputs:

• Image Inputs (Optional, 5 slots)

  • When connected, creates an image batch sized to Total_Frames for IP Adapter
  • Includes crossfade transitions controlled by Transition_Frames parameter

• Total_Frames (INT)

  • Defines total number of frames to process
  • Default: 96, Range: 16-2000

• Image_startPoint_XX (INT, XX: 01-05)

  • Keyframe starting point for each image/style
  • Image_01_startPoint fixed at 0
  • Each keyframe must be larger than previous ones
  • Range: 0-1000

• Prompt_Image_XX (STRING, XX: 01-05)

  • Prompt text for each image/keyframe
  • Used in prompt scheduling

• Transition_Frames (INT)

  • Controls transition length between images
  • Affects fade mask timing and image batch output
  • Does not affect prompt timing
  • Default: 8, Range: 0-32, Step: 4
  • Creates pre & post transition points:
    • Pre-transition: Transition_Frames/2 frames before keyframe
    • Post-transition: Transition_Frames/2 frames after keyframe
  • Allows control of ControlNet & IP Adapter strength during transitions

• Fade Mask Controls (6x FLOAT)

  • fade_mask_XX_Min (XX: 01-03)
    • Minimum weight for fade mask
    • Range: 0.0-1.0, Step: 0.1
  • fade_mask_XX_Max (XX: 01-03)
    • Maximum weight for fade mask
    • Range: 0.0-1.0, Step: 0.1
  • Set Min > Max to invert mask

Outputs:

• Image_Batch_Output (IMAGE)

  • Generated when images are connected
  • Batch sized to Total_Frames
  • Each image spans its keyframe duration
  • Includes crossfade transitions between images

• Text_Prompt_Timing (STRING)

  • Formatted timing string for prompt schedule node

• fade_mask_timing_XX (3x STRING, XX: 01-03)

  • Formatted timing strings for fade mask advanced node
  • Controls transition strength and mask behavior

Creating Loops:

For img2video loops:

1. Set last Image_startPoint equal to Total_Frames
2. Use identical first and last Prompt_Image and Image inputs

Note:

Transition system uses three control points (Pre-transition, Keyframe, Post-transition) for smooth blending of ControlNet and IP Adapter weights during transitions.

I'll create a comprehensive markdown documentation for the DP_Image_Slide_Show node:

DP_Image_Slide_Show

ComfyUI node for creating image sequences with blend mode transitions. Ideal for GIFs and videos.

• Inputs:

  • 5*optional image inputs - when connected, creates an image batch with transitions between images
  • INT - Total_Frames - how many frames will be processed (16-2000)
  • 5*INT - Image_startPoint 01-05 - the start key frame for each image, image_01_startPoint is set to 0. *notice that each keyFrame has to be bigger than the previous ones
  • INT - Transition_Frames - the number of frames for transition between images (0-32, step: 4)
  • COMBO - blend_mode - transition effect between images:
    • Normal Blend - Dissolve - Overlay - Multiply - Screen - Soft Light

• Output:

  • Image - Image_Batch_Output - exports image batch in the size of Total_Frames, with selected blend mode transitions between images

*for seamless loops, set the last Image_startPoint same as Total_Frames and use identical first/last images

*for seamless loops, set the last Image_startPoint same as Total_Frames and use identical first/last images

DP_Logo_Animator

DP_Video_Effect_Sender & DP_Video_Effect_Receiver

DP_Video_Flicker

DP_Video_Transition

DP_Smart_Saver

DP_Big_Letters

DP_Broken_Token

DP_Clean_Prompt

DP_Create_JSON

DP_Crazy_Prompt

DP_Image_Color_Analyzer

DP_Fast_Slow_Motion

DP_Five_Lora:

DP_Five_Lora_Random

DP_Float_0_1 | DP_2Floats_0_1 | DP_3Floats_0_1

DP_Image_Empty_Latent_Switch

DP_Image_Loader_Small

DP_Image_Loader_Medium

DP_Image_Loader_Big

DP_Prompt_Styler

DP_Quick_Link

DP_Random_Char

DP_Random_MinMax

DP_Aspect_Ratio

Installation

1. Clone this repository into your ComfyUI custom_nodes folder:

# Standard library imports
import os

# Third-party imports
import numpy as np
import torch
from PIL import Image

# Local imports
from .image_effects import ImageEffects