pipe-remove-background

pipe-remove-background.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#https://learn.microsoft.com/en-us/azure/ai-services/computer-vision/how-to/call-analyze-image-40?pivots=programming-language-csharp\n",
    "#https://learn.microsoft.com/en-us/azure/ai-services/computer-vision/how-to/background-removal\n",
    "#https://github.com/danielgatis/rembg/blob/main/USAGE.md"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from azure.ai.vision.imageanalysis import ImageAnalysisClient\n",
    "from azure.ai.vision.imageanalysis.models import VisualFeatures\n",
    "from azure.core.credentials import AzureKeyCredential\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "endpoint = \"https://cv-XXX.cognitiveservices.azure.com/\"\n",
    "credential = AzureKeyCredential(\"XXX\")\n",
    "\n",
    "image_file = 'truth_set/updated_2024-12-11T07_25_05_359727.jpg'\n",
    "with open(image_file, \"rb\") as f:\n",
    "    image_data = f.read()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'modelVersion': '2023-10-01', 'captionResult': {'text': 'a white pipe with a black pipe', 'confidence': 0.5815051794052124}, 'metadata': {'width': 3024, 'height': 4032}, 'readResult': {'blocks': []}}\n"
     ]
    }
   ],
   "source": [
    "client = ImageAnalysisClient(endpoint=endpoint,credential=credential)\n",
    "visual_features=[VisualFeatures.CAPTION, VisualFeatures.READ]\n",
    "result = client.analyze(image_data=image_data,visual_features=visual_features,gender_neutral_caption=True,language=\"en\",)\n",
    "print(result)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "client = ImageAnalysisClient(endpoint=endpoint,credential=credential)\n",
    "visual_features=[VisualFeatures.OBJECTS]\n",
    "result = client.analyze(image_data=image_data,visual_features=visual_features,gender_neutral_caption=True,language=\"en\",)\n",
    "\n",
    "if result.objects is not None:\n",
    "    for object in result.objects.list:\n",
    "        print(f\"   '{object.tags[0].name}', {object.bounding_box}, Confidence: {object.tags[0].confidence:.4f}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "#backgroundRemoval\n",
    "# Background removal is only available through direct REST API calls. It is not available through the SDKs.\n",
    "# Being Removed Jan 2025\n",
    "\n",
    "import requests\n",
    "\n",
    "endpoint = \"https://cv-XXX.cognitiveservices.azure.com/\"\n",
    "api_version = \"2023-02-01-preview\"\n",
    "mode = \"backgroundRemoval\"\n",
    "key = \"XXX\"\n",
    "url = \"{}computervision/imageanalysis:segment?api-version={}&mode={}\".format(endpoint, api_version, mode)\n",
    "headers = {\"Ocp-Apim-Subscription-Key\": key,\"Content-Type\": \"application/octet-stream\"}\n",
    "body = image_data\n",
    "\n",
    "response = requests.post(url, headers=headers, data=body)\n",
    "image=response.content\n",
    "with open(\"output/pipe_nobackground.png\", \"wb\") as file:\n",
    "    file.write(image)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading data from 'https://github.com/danielgatis/rembg/releases/download/v0.0.0/u2net.onnx' to file 'C:\\Users\\dixonjames\\.u2net\\u2net.onnx'.\n",
      "100%|########################################| 176M/176M [00:00<00:00, 174GB/s]\n"
     ]
    }
   ],
   "source": [
    "from rembg import remove\n",
    "from PIL import Image\n",
    "import io\n",
    "\n",
    "output_data = remove(image_data)\n",
    "output_image_path = 'output/pipe_no_bg_rembg.png'\n",
    "output_image = Image.open(io.BytesIO(output_data))\n",
    "output_image.save(output_image_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "ename": "ModuleNotFoundError",
     "evalue": "No module named 'rembg.session'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[16], line 5\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;66;03m# To maintain a smooth operation of your models, install the open-source Florence 2 model and use its Region \u001b[39;00m\n\u001b[0;32m      2\u001b[0m \u001b[38;5;66;03m# to segmentation feature, which allows for a similar background removal operation.\u001b[39;00m\n\u001b[0;32m      4\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mrembg\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m remove\n\u001b[1;32m----> 5\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mrembg\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01msession\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m new_session\n\u001b[0;32m      6\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mPIL\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m Image\n\u001b[0;32m      7\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mio\u001b[39;00m\n",
      "\u001b[1;31mModuleNotFoundError\u001b[0m: No module named 'rembg.session'"
     ]
    }
   ],
   "source": [
    "# To maintain a smooth operation of your models, install the open-source Florence 2 model and use its Region \n",
    "# to segmentation feature, which allows for a similar background removal operation.\n",
    "#Note - could not find this session class in the SDK\n",
    "\n",
    "from rembg import remove\n",
    "from rembg.session import new_session\n",
    "from PIL import Image\n",
    "import io\n",
    "\n",
    "session = new_session(model_name='florence2')  # Replace 'u2netp' with 'florence2' if the model is renamed.\n",
    "output_data = remove(image_data, session=session)\n",
    "output_image_path = 'output/pipe_no_bg_florence2.png'\n",
    "output_image = Image.open(io.BytesIO(output_data))\n",
    "output_image.save(output_image_path)\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}