BLIP image captioning
This experimental node converts an image represented as colored points into a textual description of its content. The description is stored as a point attribute called 'prompt'.
RemBG
This node extracts and replaces background of any image with solid color.
CLIPseg Image Mask
This node extracts low resolution mask from an input image based on another image when it is difficult to describe in text what exactly needs to be identified in the image.
CLIPseg Prompt Mask
This node extracts low resolution mask from an input image based on text prompt. The mask can be used to isolate specific areas of interest for further processing downstream.
Semantic Similarity
This node takes two textual prompts as input and returns a Similarity attribute between 0 and 1 indicating how closely related the prompts are in meaning.
SD Image Roll
This node shifts or offsets an input pixels along a specified axis.
SD Image Tiling
This node repeats images along a specified axis with preserving image dimensions.
SD Shapes to Points
This node generates black and white mask from different kinds of shapes with ability to blur edges.
SD COP2 Processor
This node contains COP2 subnetwork for image manipulations and outputs colored points.
SD Points to HeightField
This node converts colored points channels to heightfield volumes Height and Mask.
SD HeightField to Points
This node converts heightfield volumes volumes Height or Mask to colored points. It also has COP2 subnetwork inside for image post-processing.
SD Image Python
This node allows processing Colored Points by representing them as a three-dimensional numpy array in the variable "img" with a shape of (w, h, c), where w and h are the width and height in pixels, "c" - represents r,g,b channels.
SD MatPlotLib Python
This node generates an image from a Matplotlib figure. You can customize the figure parameters and use gridspec if desired to layout plots.
Experimental nodes for Machine Learning Operators in Houdini distributed under a Limited Commercial (Indie) license.
Machine Learning Operators (MLOPs) is a collection of nodes that empower Houdini with the capabilities of machine learning. By establishing a framework for working with image data at the SOP (Points) context and streamlining the integration of Python dependencies, MLOPs opens up exciting possibilities. While earlier versions of MLOPs primarily focused on working with Stable Diffusion models, I took it upon myself to enhance its functionality by introducing new nodes that cater to the diverse needs of artists when constructing MLOPs-driven pipelines.
These additional utility nodes, designed to seamlessly integrate within the MLOPs ecosystem, have garnered the attention of MLOPs developers. Many of them are slated for inclusion in the comprehensive commercial release, while others have already found their place in it. Nevertheless, I believe in the power of sharing knowledge and empowering the community. As a result, I have chosen to make these nodes available to the public, accessible through my GitHub repository. My hope is that they will prove valuable to those who wish to explore the possibilities of MLOps and elevate their artistic endeavors.
On the dedicated GitHub page: https://github.com/Faitel/MLOPsLC you'll find a comprehensive repository with all assets, documentation, installation guide, node's descriptions and other useful information.
Here are just a few examples of the nodes available on Github:
