[New Notebook] Biological Texture Generation (Iris) using LCM & OpenVINO

notebooks/biological-texture-generation-lcm/README.md

@@ -0,0 +1,14 @@
+# Biological Texture Generation (Iris) using LCM & OpenVINO™
+
+This notebook demonstrates how to generate high-fidelity biological textures, specifically human iris patterns, using **Latent Consistency Models (LCM)**.
+
+## Key Features
+- **Model:** [SimianLuo/LCM_Dreamshaper_v7](https://huggingface.co/SimianLuo/LCM_Dreamshaper_v7)
+- **Library:** [Optimum Intel](https://github.com/huggingface/optimum-intel) with OpenVINO™ backend.
+- **Performance:** Generates high-quality images in just 4-8 steps (vs 25-50 steps for standard Stable Diffusion).
+- **Hardware:** Supports Intel CPU, GPU (iGPU/dGPU), and NPU.
+
+## Usage
+1. Install dependencies.
+2. Select your target device (CPU/GPU) from the widget.
+3. Run the pipeline to generate synthetic iris data for biometric research.

notebooks/biological-texture-generation-lcm/biological-texture-generation-lcm.ipynb

@@ -0,0 +1,106 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/"
+    },
+    "id": "cXoFhlAHytJW",
+    "outputId": "aa7ff889-4d2e-4ee9-ae63-385037be013a"
+   },
+   "outputs": [],
+   "source": [
+    "%pip install -q \"openvino>=2023.2.0\" \"git+https://github.com/huggingface/optimum-intel.git\" \"diffusers\" \"transformers\" \"gradio\" --extra-index-url https://download.pytorch.org/whl/cpu"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/"
+    },
+    "id": "4hhdDBTlOHhv",
+    "outputId": "850624c3-4a1e-42ea-b464-cb523b0927e9"
+   },
+   "outputs": [],
+   "source": [
+    "import ipywidgets as widgets\n",
+    "import openvino as ov\n",
+    "from optimum.intel.openvino import OVLatentConsistencyModelPipeline\n",
+    "import gc\n",
+    "\n",
+    "# Initialize OpenVINO Core\n",
+    "core = ov.Core()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Create a dropdown widget to select the device (CPU, GPU, AUTO)\n",
+    "device = widgets.Dropdown(\n",
+    "    options=core.available_devices + [\"AUTO\"],\n",
+    "    value='AUTO',\n",
+    "    description='Device:',\n",
+    "    disabled=False,\n",
+    ")\n",
+    "\n",
+    "device"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "colab": {
+     "base_uri": "https://localhost:8080/"
+    },
+    "id": "UslPhY2mRFCD",
+    "outputId": "98b561d7-1460-47a7-8950-d31e9aa22130"
+   },
+   "outputs": [],
+   "source": [
+    "model_id = \"SimianLuo/LCM_Dreamshaper_v7\"\n",
+    "\n",
+    "print(f\"Loading {model_id} and converting to OpenVINO IR... (This may take a few minutes)\")\n",
+    "\n",
+    "# Load the model and export it to OpenVINO format automatically\n",
+    "pipe = OVLatentConsistencyModelPipeline.from_pretrained(\n",
+    "    model_id,\n",
+    "    export=True,\n",
+    "    device=device.value \n",
+    ")\n",
+    "\n",
+    "# Compile the model explicitly for the selected device\n",
+    "pipe.compile()\n",
+    "print(f\"✅ Model successfully compiled for {device.value}\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "4JHQFlfniXhk"
+   },
+   "source": []
+  }
+ ],
+ "metadata": {
+  "colab": {
+   "provenance": []
+  },
+  "kernelspec": {
+   "display_name": "Python 3",
+   "name": "python3"
+  },
+  "language_info": {
+   "name": "python"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
+}