openai-cookbook/examples/dalle/How_to_create_dynamic_masks_with_DALL-E_and_Segment_Anything.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NCiCQizqxBSv"
      },
      "source": [
        "# How to create dynamic masks with DALL·E and Segment Anything\n",
        "\n",
        "Segment Anything is a model from Meta that can be used to select portions of images. Combined with DALL·E's ability to inpaint specified portions of images, you can use Segment Anything to easily select any part of an image you'd like to alter.\n",
        "\n",
        "In this notebook, we'll use these tools to become fashion designers and dynamically replace our digital models' outfits with tailored, original creations. The notebook follows this flow:\n",
        "- **Setup:** Initialise your libraries and any location directories.\n",
        "- **Generate original image:** Make an original image that we'll create dynamic masks from.\n",
        "- **Generate mask:** Use Segment Anything to create a dynamic mask.\n",
        "- **Create new image:** Generate a new image with the masked area inpainted with a fresh prompt."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "91E6gN8cxE4e"
      },
      "source": [
        "## Setup\n",
        "\n",
        "To get started we'll need to follow the [instructions](https://github.com/facebookresearch/segment-anything) for using the Segment Anything (SAM) model open-sourced by Meta. As of May 2023, the key steps are:\n",
        "- Install [Pytorch](https://pytorch.org/get-started/locally/) (version 1.7+).\n",
        "- Install the library using ```pip install git+https://github.com/facebookresearch/segment-anything.git```.\n",
        "- Install dependencies using ```pip install opencv-python pycocotools matplotlib onnxruntime onnx```.\n",
        "- Download a [model checkpoint](https://github.com/facebookresearch/segment-anything#model-checkpoints) to use (default size is 2.4 GB).\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "0Qs6pVTvtZIg"
      },
      "outputs": [],
      "source": [
        "!pip install torch torchvision torchaudio\n",
        "!pip install git+https://github.com/facebookresearch/segment-anything.git\n",
        "!pip install opencv-python pycocotools matplotlib onnxruntime onnx\n",
        "!pip install requests\n",
        "!pip install openai\n",
        "!pip install numpy"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "vQC_oUx8tqS1"
      },
      "outputs": [],
      "source": [
        "!wget https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "wBIC4srntdM8"
      },
      "outputs": [],
      "source": [
        "import cv2\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.image as mpimg\n",
        "from matplotlib import rcParams\n",
        "import numpy as np\n",
        "from openai import OpenAI\n",
        "import os\n",
        "from PIL import Image\n",
        "import requests\n",
        "from segment_anything import sam_model_registry, SamAutomaticMaskGenerator, SamPredictor\n",
        "import torch\n",
        "\n",
        "# Set directories for generation images and edit images\n",
        "base_image_dir = os.path.join(\"images\", \"01_generations\")\n",
        "mask_dir = os.path.join(\"images\", \"02_masks\")\n",
        "edit_image_dir = os.path.join(\"images\", \"03_edits\")\n",
        "\n",
        "# Point to your downloaded SAM model\n",
        "sam_model_filepath = \"./sam_vit_h_4b8939.pth\"\n",
        "\n",
        "# Initiate SAM model\n",
        "sam = sam_model_registry[\"default\"](checkpoint=sam_model_filepath)\n",
        "\n",
        "# Initiate openAI client\n",
        "client = OpenAI(api_key=os.environ.get(\"OPENAI_API_KEY\", \"<your OpenAI API key if not set as env var>\"))\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6UzsByKRxHZt"
      },
      "source": [
        "## Generate original image\n",
        "\n",
        "First we'll create an original image which we'll generate masks from."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "KRxf_PHwuDDb"
      },
      "outputs": [],
      "source": [
        "def process_dalle_images(response, filename, image_dir):\n",
        "    # save the images\n",
        "    urls = [datum.url for datum in response.data]  # extract URLs\n",
        "    images = [requests.get(url).content for url in urls]  # download images\n",
        "    image_names = [f\"{filename}_{i + 1}.png\" for i in range(len(images))]  # create names\n",
        "    filepaths = [os.path.join(image_dir, name) for name in image_names]  # create filepaths\n",
        "    for image, filepath in zip(images, filepaths):  # loop through the variations\n",
        "        with open(filepath, \"wb\") as image_file:  # open the file\n",
        "            image_file.write(image)  # write the image to the file\n",
        "\n",
        "    return filepaths"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "937VVpQ7uH7M"
      },
      "outputs": [],
      "source": [
        "dalle_prompt = '''\n",
        "Full length, zoomed out photo of our premium Lederhosen-inspired jumpsuit.\n",
        "Showcase the intricate hand-stitched details and high-quality leather, while highlighting the perfect blend of Austrian heritage and modern fashion.\n",
        "This piece appeals to a sophisticated, trendsetting audience who appreciates cultural fusion and innovative design.\n",
        "'''"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "b5nI3PGvuJV6"
      },
      "outputs": [],
      "source": [
        "# Generate your images\n",
        "generation_response = client.images.generate(\n",
        "    model = \"dall-e-3\",\n",
        "    prompt=dalle_prompt,\n",
        "    n=3,\n",
        "    size=\"1024x1024\",\n",
        "    response_format=\"url\",\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "id": "e6CclNUZuKdb"
      },
      "outputs": [],
      "source": [
        "filepaths = process_dalle_images(generation_response, \"generation\", base_image_dir)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "aaJaZWxSuMXG"
      },
      "outputs": [],
      "source": [
        "# print the new generations\n",
        "for filepath in filepaths:\n",
        "    print(filepath)\n",
        "    display(Image.open(filepath))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gbnNN_GuxLJP"
      },
      "source": [
        "## Generate Mask\n",
        "\n",
        "Next we'll load up one of our images and generate masks.\n",
        "\n",
        "For this demonstration we're picking a UX where we \"click\" on a point on the image to generate masks from. However, there are [example notebooks](https://github.com/facebookresearch/segment-anything/blob/main/notebooks/automatic_mask_generator_example.ipynb) provided by Meta which show how to generate every possible mask for an image, draw a box, and some other useful approaches."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "gGiFFn1Kux6_"
      },
      "outputs": [],
      "source": [
        "# Pick one of your generated images\n",
        "chosen_image = \"images/01_generations/generation_2.png\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "id": "94XUO7R-u0Nj"
      },
      "outputs": [],
      "source": [
        "# Function to display mask using matplotlib\n",
        "def show_mask(mask, ax):\n",
        "    color = np.array([30 / 255, 144 / 255, 255 / 255, 0.6])\n",
        "    h, w = mask.shape[-2:]\n",
        "    mask_image = mask.reshape(h, w, 1) * color.reshape(1, 1, -1)\n",
        "    ax.imshow(mask_image)\n",
        "\n",
        "\n",
        "# Function to display where we've \"clicked\"\n",
        "def show_points(coords, labels, ax, marker_size=375):\n",
        "    pos_points = coords[labels == 1]\n",
        "    neg_points = coords[labels == 0]\n",
        "    ax.scatter(\n",
        "        pos_points[:, 0],\n",
        "        pos_points[:, 1],\n",
        "        color=\"green\",\n",
        "        marker=\"*\",\n",
        "        s=marker_size,\n",
        "        edgecolor=\"white\",\n",
        "        linewidth=1.25,\n",
        "    )\n",
        "    ax.scatter(\n",
        "        neg_points[:, 0],\n",
        "        neg_points[:, 1],\n",
        "        color=\"red\",\n",
        "        marker=\"*\",\n",
        "        s=marker_size,\n",
        "        edgecolor=\"white\",\n",
        "        linewidth=1.25,\n",
        "    )\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 835
        },
        "id": "DBdLivJcu1Ha",
        "outputId": "15792f47-5bac-423f-bbd0-5053e9f480e9"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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
            "text/plain": [
              "<Figure size 1000x1000 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Load chosen image using opencv\n",
        "image = cv2.imread(chosen_image)\n",
        "image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n",
        "\n",
        "# Display our chosen image\n",
        "plt.figure(figsize=(10, 10))\n",
        "plt.imshow(image)\n",
        "plt.axis(\"on\")\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 835
        },
        "id": "YWaGwMXhu2GS",
        "outputId": "9bc19927-db97-4ce8-e06e-f9a454dcf44a"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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
            "text/plain": [
              "<Figure size 1000x1000 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Set the pixel coordinates for our \"click\" to assign masks\n",
        "input_point = np.array([[525, 325]])\n",
        "input_label = np.array([1])\n",
        "\n",
        "# Display the point we've clicked on\n",
        "plt.figure(figsize=(10, 10))\n",
        "plt.imshow(image)\n",
        "show_points(input_point, input_label, plt.gca())\n",
        "plt.axis(\"on\")\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "H8IRnYNPu270",
        "outputId": "531b4cb1-af6e-4d87-fc24-e871e9c83fc0"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(3, 1024, 1024)"
            ]
          },
          "execution_count": 14,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Initiate predictor with Segment Anything model\n",
        "predictor = SamPredictor(sam)\n",
        "predictor.set_image(image)\n",
        "\n",
        "# Use the predictor to gather masks for the point we clicked\n",
        "masks, scores, logits = predictor.predict(\n",
        "    point_coords=input_point,\n",
        "    point_labels=input_label,\n",
        "    multimask_output=True,\n",
        ")\n",
        "\n",
        "# Check the shape - should be three masks of the same dimensions as our image\n",
        "masks.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "ZpMbl7-Ku4IL",
        "outputId": "5b039f47-eca3-4c3b-e25c-275f62336827"
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAxYAAAMxCAYAAAB2FyUpAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOy9efSlWVUe/Ozz3nt/Q01d1Q0NNC1EEZUFMSgSUWJjCwiiolki4oTiFDUaBEVdshQlUXFFkDgkGhQwS0DiQNDlBAIKDlHMpyAEGhS0G6TpsYbfeO979vfH3s/e+62eqmgDDvdoU1X33vc95+zh2dMZRFUV67Zu67Zu67Zu67Zu67Zu67Zud6O1j/QA1m3d1m3d1m3d1m3d1m3d1u0ff1sHFuu2buu2buu2buu2buu2but2t9s6sFi3dVu3dVu3dVu3dVu3dVu3u93WgcW6rdu6rdu6rdu6rdu6rdu63e22DizWbd3Wbd3Wbd3Wbd3Wbd3W7W63dWCxbuu2buu2buu2buu2buu2bne7rQOLdVu3dVu3dVu3dVu3dVu3dbvbbR1YrNu6rdu6rdu6rdu6rdu6rdvdbuvAYt3Wbd3Wbd3Wbd3Wbd3Wbd3udlsHFuu2buv2z6Y96lGPgojgOc95zkd6KOu2buu2buu2bv/k2jqwWLd1W7e7bM95znMgIvHfK17xirt85glPeMLkmfe+973/7wf6EWzveMc78HM/93P45m/+ZjziEY/A9vZ2zP3D1d7+9rfjW7/1W/GJn/iJOHHiBBaLBe5zn/vgoQ99KL7sy74M/+2//Tdcc801H7bx/FNq119/PZ75zGfi4z7u47C1tYVTp07h3/ybf4MXvehFUNW79e5xHPE//sf/wGMe8xhcdtll2NjYwH3ve1885SlPwR/90R/d6bO/8Au/gG/91m/FIx/5SNz//vfHkSNHsLm5iY/6qI/CF37hF+KXf/mX7/T5+9///hM9vb3/HvnIR96t+a3buq3bP6Om67Zu67Zud9G+7/u+TwHEf4997GPv9Pfve9/7dBiGyTPvec97PjyDvZN21VVXKQD9vu/7vv9n7769/z4c7Ud+5Ed0NptN+r3kkkt0a2tr8tlVV131YRnPP6X25je/WS+99NKg4dGjRye0/uzP/mw9ODj4kN599uxZffSjHx3vGoZBT548qa01BaCtNf2RH/mRO3z+yJEjE/6eOHFCF4vFbfT13Llzt/v8/e53PwWgx48f18svv/x2//v8z//8D2lu67Zu6/bPr60rFuu2but2we2yyy7DkSNH8NrXvhbXXXfdHf7u53/+5zGOI+5///t/+Ab3EW6z2QwPetCD8OVf/uV4/vOfj2c84xkftr5/5Vd+Bc961rOwWq3wGZ/xGfid3/kd7O3t4ZZbbsHu7i6uu+46vPzlL8cXfdEXYbFYfNjG9U+hnT59Gp/7uZ+Lm266CR//8R+PP/3TP8XZs2exs7ODn/iJn8B8Psdv//Zv4+lPf/qH9P6v+7qvw2tf+1q01vCDP/iDuOWWW3DzzTfjxhtvxLOe9Sz03vGsZz0Lr371q2/3+a/92q/FS1/6UrzrXe/C/v4+br31Vuzv7+Ov/uqv8C3f8i0AgN/5nd/BM5/5zDsdxwtf+EJ84AMfuN3//tf/+l8f0tzWbd3W7Z9h+0hHNuu2buv2D7+xYnG/+91Pn/rUpyoAfe5zn3uHv3/gAx+oAPQ5z3nOP5uKxWq1mvz7xS9+8YetYvFpn/ZpCkAf/OAH63K5vNPf7u7u/j8fzz+l9uxnP1sB6NbWlv71X//1bb7/wR/8wag0vPOd77yod7/lLW8JGXn6059+u7958pOfrAD0AQ94gI7jeNHjf8pTnhLjPzw8vM33rFi8+MUvvuh3r9u6rdu6nd/WFYt1W7d1u6j21V/91QCAl7zkJbf7/Zve9CZcc801+OiP/mh8xmd8xp2+6y//8i/xnOc8B1dffTU+5mM+BltbWzh+/Dge+tCH4tnPfjZuvPHGO3x2tVrhZ37mZ/CoRz0Kl112GebzOS699FJ83Md9HJ785CfjZ3/2Zy96bi996Usxn88hIvie7/mei3p2GIaL7u/vq/35n/85AOBzPudzMJvN7vS3W1tbd/jdzs4Onv/85+Oqq67CZZddhsVigfve97646qqr8KM/+qO4/vrrb/e5N7zhDXjSk56EK664AhsbG7jsssvwWZ/1WXjxi1+McRxv9xnu23nUox4FAPjlX/5lPPaxj8U973lPtNZus8H+hhtuwLOf/Ww89KEPxYkTJ7C5uYmP/uiPxtd8zdfgbW97253O+e60n//5nwcAfMmXfAn+xb/4F7f5/lu+5Vtw9OhRjOOIX/iFX7iod//Gb/xG/P07vuM7bvc3z3rWswAA7373u/GmN73pot4PAJ/6qZ8KANjb28PNN9980c+v27qt27pdVPtIRzbrtm7r9g+/1YpF710/5mM+RgHo7/3e793mt0972tMUgP7AD/yAvv71r7/TigWzpQB0c3NTT506pSISn11xxRX6jne84zbPrVYrfcxjHnObteUbGxt3urfhzioWP/RDPxRr2n/iJ37iQ6JTbR/OisX29rYC0C/90i/9kN/xZ3/2Z3rllVfGmFtreurUqQlNX/CCF9zmuW/7tm+L70VEL7nkksn+mquvvlrPnDlzm+coU1dddZU+4xnPiOdPnjypwzBMePSa17xGL7nkknjnfD6f7C1YLBb60pe+9HbnxQrbh8KHd7zjHfHsK1/5yjv83eMf/3gFoJ/6qZ96Ue//xm/8xpDdO2o7OzuhE9/zPd9zUe9XVf2iL/qi2BfSe7/N9+uKxbqt27r9fbZ1xWLd1m3dLqqJCL7qq74KAPBzP/dzk+92dnbwyle+Eq21+M2dtauuugoveclL8Dd/8zfY29vDTTfdhP39fbz2ta/Fwx/+cLzvfe/Dl37pl97muZe//OV4zWteg83NTbzoRS/C2bNnceutt2Jvbw/XX389fuVXfgVf9EVfdEHzUVX8h//wH/Dd3/3d2NjYwC/+4i/im7/5my/o2X8o7eEPfzgA4JWvfCVe9rKXofd+Uc9fe+21+OzP/mxce+21uPLKK/GKV7wCZ8+exU033YS9vT287W1vw3Oe8xzc4x73mDz3Ez/xE3jBC14AAPj6r/96vP/978ctt9yC06dP4wUveAFmsxle97rX4eu+7uvusO8/+7M/w/Of/3x853d+J66//nrcfPPN2NnZicrYW9/6Vnz+538+br31Vnzd130d3v72t2Nvbw/nzp3D3/zN3+CbvumbcHh4iK/5mq/Bm9/85oua9121v/zLv4y/P/jBD77D3/G7t7/97R9SP3fGr957nDr11re+9YLed/r0abz5zW/G0572NPzSL/0SAOAZz3jGnZ5Q9p//83/GFVdcgcVigVOnTuGRj3wkfviHfxi33HLLRcxk3dZt3f7Zt490ZLNu67Zu//BbrVioqv7t3/6tttb0yJEjevbs2fjdz/3czykAfcxjHqOqepcViztrZ8+e1csvv1wB6Bvf+MbJd8z0fv3Xf/1FvfP8isXBwYF+8Rd/cWSNX//611/U++6sfTgrFm94wxsmpxTd61730i/+4i/WH/mRH9HXve51d3giENuXf/mXKwC99NJL9W//9m8vqM/d3V09deqUAtCnPOUpt/ub//Jf/kuM6c1vfvPku3rS2DOe8Yw77Ofqq69WAPrd3/3dd/ibb/3Wb1UA+sQnPvE2392dikUd/+nTp+/wdz/2Yz8Wv6v6cFftec97Xjz33ve+93Z/80d/9Efxm4c97GF3+K6Xv/zlt3si2cbGhn7Xd33XbfYAsZ1fNayVIcrSm970pgue07qt27r9827risW6rdu6XXS78sor8ehHPzoqFGwvfvGLAQBPe9rT7nYfR48exVVXXQUAt1lbfskllwAAPvCBD3zI7z9z5gwe97jH4ZWvfCXufe974/d///djvf8/tnbVVVfht37rt/BxH/dxAIwur3zlK/GsZz0LV199NU6ePIknPOEJ+P3f//3bPLuzs4Nf/MVfBAB813d9F6688soL6vM1r
            "text/plain": [
              "<Figure size 1000x1000 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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
            "text/plain": [
              "<Figure size 1000x1000 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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
            "text/plain": [
              "<Figure size 1000x1000 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Display the possible masks we can select along with their confidence\n",
        "for i, (mask, score) in enumerate(zip(masks, scores)):\n",
        "    plt.figure(figsize=(10, 10))\n",
        "    plt.imshow(image)\n",
        "    show_mask(mask, plt.gca())\n",
        "    show_points(input_point, input_label, plt.gca())\n",
        "    plt.title(f\"Mask {i+1}, Score: {score:.3f}\", fontsize=18)\n",
        "    plt.axis(\"off\")\n",
        "    plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "id": "pUfL0bvTu6cf"
      },
      "outputs": [],
      "source": [
        "# Choose which mask you'd like to use\n",
        "chosen_mask = masks[1]\n",
        "\n",
        "# We'll now reverse the mask so that it is clear and everything else is white\n",
        "chosen_mask = chosen_mask.astype(\"uint8\")\n",
        "chosen_mask[chosen_mask != 0] = 255\n",
        "chosen_mask[chosen_mask == 0] = 1\n",
        "chosen_mask[chosen_mask == 255] = 0\n",
        "chosen_mask[chosen_mask == 1] = 255\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "_GAYz_WTu7rr",
        "outputId": "b94b1fa4-1764-4149-f828-ee955a813dde"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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
            "text/plain": [
              "<PIL.Image.Image image mode=RGBA size=1024x1024>"
            ]
          },
          "execution_count": 17,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# create a base blank mask\n",
        "width = 1024\n",
        "height = 1024\n",
        "mask = Image.new(\"RGBA\", (width, height), (0, 0, 0, 1))  # create an opaque image mask\n",
        "\n",
        "# Convert mask back to pixels to add our mask replacing the third dimension\n",
        "pix = np.array(mask)\n",
        "pix[:, :, 3] = chosen_mask\n",
        "\n",
        "# Convert pixels back to an RGBA image and display\n",
        "new_mask = Image.fromarray(pix, \"RGBA\")\n",
        "new_mask"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "id": "ddzlWgXLu9kj"
      },
      "outputs": [],
      "source": [
        "# We'll save this mask for re-use for our edit\n",
        "new_mask.save(os.path.join(mask_dir, \"new_mask.png\"))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cIyQtNInxQsu"
      },
      "source": [
        "## Create new image\n",
        "\n",
        "Now we'll combine our original image with the mask and the Edit endpoint for DALLE to inpaint the transparent area according to a new prompt. (as 0f January 2024 dall-e-2 is the only model that supports edits)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "id": "S-aVmhj8u_1W"
      },
      "outputs": [],
      "source": [
        "# edit an image\n",
        "edit_response = client.images.edit(\n",
        "    image=open(chosen_image, \"rb\"),  # from the generation section\n",
        "    mask=open(os.path.join(mask_dir, \"new_mask.png\"), \"rb\"),  # from right above\n",
        "    prompt=\"Brilliant leather Lederhosen with a formal look, detailed, intricate, photorealistic\",  # provide a prompt to fill the space\n",
        "    n=3,\n",
        "    size=\"1024x1024\",\n",
        "    response_format=\"url\",\n",
        ")\n",
        "\n",
        "edit_filepaths = process_dalle_images(edit_response, \"edits\", edit_image_dir)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 304
        },
        "id": "3wF9JzR1vCbV",
        "outputId": "1927179c-9c0d-4b2d-c239-c00e30c01b01"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x791b1f4c58a0>"
            ]
          },
          "execution_count": 21,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2gAAAEOCAYAAAD8JDRTAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOz9d9BmS37fh326T3jiGyfP3DT33sUGAAsQmQAYABBgkFQMlmXQLKto0Sr6D/MfURZtslwsV7lcJUty2VJJVbLLYrBLtKmCCRACxGQAu4sFsAG7WOzee/fGmbl38sz7zpuecFK3/+gTuvuc550ZcO+EF+c3Ne9zQodf/PY5fToIrbWmp5566qmnnnrqqaeeeuqpp6dO8mkz0FNPPfXUU0899dRTTz311JOh/gWtp5566qmnnnrqqaeeeurpGaH+Ba2nnnrqqaeeeuqpp5566ukZof4Fraeeeuqpp5566qmnnnrq6Rmh/gWtp5566qmnnnrqqaeeeurpGaH+Ba2nnnrqqaeeeuqpp5566ukZof4Fraeeeuqpp5566qmnnnrq6Rmh/gWtp5566qmnnnrqqaeeeurpGaHwURMWRQECNCB0+QsIAVoLzBVDurrrXm6RKP9orRHmDC3Ke7rKau7pqsKHUFNmx3Uavh+HdFmA6CrYLrxO3Fz366vl0Nb1UocCjRbm12HWKxO9Qg4rrX1fAwd3bnL/S/+SyXKPMB7CYIicTCAeosOY5eEB96++z803v8k7H1zla1dvsMgUL5zb5sUL5xhP1/jMz/2P+NTP/HkGg5hxKJDSCCgqpdd6atQkKG2qdX3iyu1pWohahrpYTVVJS/+1bbSpp6WuDh1qSzvCulrVWlf3CFu4d3BD6Skrs9ciYdtJ13X7aexI6Mpfndmy6NIQQvs1tH3Dd98wCB4i9fNBPWb1mNVjVpt6zHp2qcesHrN6zGrTH1bMeuQXtNovPKNq+6IvtXYZtIqpbiN0I7JGl0WZAsz1Ksra4d6lBPegAraWGG1mrMK0dtPVBtLOWdtL/QCncVxteXgFipWjGz51XUmXMU3uRpjjwK8C3yZ4BevnLqK+/yfJ3v0aYSiRgwEqHqIQpMslO3fvcOvqFT746AZv3rzH7b0Z57bXuXh6m9EgIh4NydMlozgkCiWB9ASs1GH7hyWjRjRyW4kb25cXLVvUOrcK9k1muVrthi2H8BsRIdz0VqamOo2wQMxJ5vt5dWilb4NGpYVu4NIti9tAUOXtAk2B0KKVU9Q6bMqxf+2G34+XE0M9ZvWY1WNWj1nPE/WY1WNWj1k9ZpX06C9obU2UtbcDSWhfxFUA4oZIpUpfbf6ZnctRgs0DZQDVwKRdo/vMVAb3GK8cv4G2FeDTEtI6qfnRblK/LiidwNOni1xWfRq0aHiyAdAjIQQbly5z7/CA5P5VwiInOXjAIs3YuX2La299i2sfXOHdj+5wY/eQQEpevXSe8+fOogXkQpJlKUIpQtEltFWt8FRrBZOwEuuW7RsA6DBNqxFw8toJy4rsxq3q+bEbM8dEomrEzAVhC+Hx5su7yjzt3LYglrk6AMg0Xr6nl+WIBoTNb9n76TjyanYcNfmJtODEUI9Z9JjVY1aPWc8R9ZhFj1k9ZvWYZejRX9AAjy/X2JRWXVWvGwsOODQGMNZuhGmFaV2Gfa8JauEqwjfyKva0e1LDVOV8ZVV1jSuCswZSoT1dlNL64tioV59qFzzagnse1xEI/rVSeBFIBudf5K3f/zLR0T3mR4c8ODjg5s1bXPvoBgeHM+7tH1IUipcvneP1V15ifWOdvf19ojBgEECeZ8DAr64tEw3gdYH+KvXZv3XI2Jkq9dplVeBq56tsobW55/iuN7yhqgP/viuhbS7f/32uReuKVY5Vte+X7nkblEWNnnaPTZO7G9KbUqreJzsOV6c/AdRjVo9ZPWb1mPU8UY9ZPWb1mNVjFo/7gqY9g9cabIOGPya1SuYDSHXgxJhTQScblrKqsoXDn+0rwlPocVS7jBMAAl0y2RjQYcYSSruVlzfqZF0RUxVVBYDQjTp9b12VF9pv6XV9utaxiEe8df0+s/e+zuGDXfaOZsxmc1RekOU5EsH62oQf+Ox38+KrrxIEkjTL0FGMRJGkKdPW91pPD7b+rF/bRbrE6brmFCEsvNRu+i59absbyPGxYxRZJxctn2mZoaWCZky0DxrdTWBb5lbDVOmtwgsLpipdCAtI/HrcdI32y9HTdQw2YPtw3TxX1GNWj1k9ZvWY9TxRj1k9ZvWY1WMWjzUHrbF8XYmmZqDi0lcY9rl2760KJG39bZlP+LI1k1ub8aCNwoSV77jAs+t2+KocsKzXHs/sCNNRj7Z47QSuVYEuaOZwVvm70nppar+wwdyjOIrQBLz7wVX29w8I0IwHIeNByGFRMB4Nef0Tr/HK5cusra+D0AyGA+ZJynL3Hnu3b7G5tY301/8s6zaqchmrDv0gehiYO0Fg2cPXhYNj1m9dnxeZD4uPVYDhseHJ0ZbGbdqwjGrx0kKNduXC+vUBzY+Pms9ST8K62MjdKMkeYrAK3J5b6jGrx6wes3rMep6ox6wes3rM6jGrpMf4gqadn1YgljU6n807x1la7Hll2IxXPSyuMG6ZmkZxlXJr9nz+VtRj83Gs4o4DgM5Cccf4PgS4qkVg6vTaVo+XWUDVvVG9z2srlfHPJqdNMpCsb6zx+uuXSedzsuUCipz9wyOiSHHm7Caf/MSrbJ3aYjwagYDtU6dJb99Cpwtuv/cG5179BNFoAMKfhqnruutJyOWn764JxDbPjmhU9ncVVjm+4yfCUt4KHa/C9i6AOC69c6+jAbNz2nXUTat2ZbbHa9v37Xp80LIbInOv26maibyW8t1DJ/eJesipqccsHna/x6weszrq6DHraVGPWTzsfo9ZPWZ11HESMeuRX9AqP6g/8FlCutqorok6ELpWcWkK7bzTmmhog4S5IVYHewcQ1PzrjvS2MlcBzkMwsEueVnGl4fwswgqT9go8mJWHcPM2QWUxqMvzBn07dRxQ8NJLLxOolCJZspgvuHHjNpNpzquf+ASXX3uN9bUpQRgh0MSDAYGQqCxleesD9u7fZ3LpIoG0dGYpSVQ8VAzrJp3vD3VPmcen02NYl2F5QRWL2vKLFcDsmGlVw8cxJrYwqQu8TDrRyOMFt1+1U5duNyqrGpnO63q1SMKygdBuWrw8jTednMeeHrP8TLQr7zGr/NtjVlcVPWY9Weoxy89Eu/Ies8q/PWZ1VXGSMOuRX9CaiZelE/sG186P4zdYSY47t/M35y5k1KXrTtvX9VbHjuPaDoubv8a9Fc7XTugxb1/3BHOcpDos9Vd/ChVu3TW/FpN+dY5I5XjqTuGsk7woUOmcQRgwGawjphPWtgTxcEI4GPLC5VfZ3N4mkgKtC9JlYvgscvLlgtHgiI++8Ttsn/63mI4iixHf2AItSjcsexc6hyR7IOLbyNWbcMSpVtVx0ou2DTv9xI/iUmcuOLdFs8tqi6Np9zxS2qY7IJ2eKdxP4ZbEnRWKrjS4JkE44nXCgnttVQA8f9Rjlp/QY77HrB6zesx6pqjHLD+hx3yPWT1m/SHCrMcb4mgL32UNyhDv8JAukFgFGn6Qd3JTOohtZN+ofvDZPPi/2j9ZIZ9TyAoe67d7vzzrvAVQNQP2FMg233Z5zsaVHeU1pTU3i6JAFjmhhDCOiQdDtFKMRmMmm1usb24ThSFCglYKmeYEUcza1hbBYMhgPOXwxjvs3LjG5PXXzITImjkb5bzt/ryGxmbXa5Ocia9+INsNUr1/RlVFd2w2rNX1lIn9Rk+7RhP1X9o9NvaxVUkXCz5oOPbUTR3NMArhlevWJbouumJ6fLo+0OLPLuckLVndY1ZbkB6zesxyZOkx69miHrNagvSY1WOWI8sfHsx6jBc00T6zma4rLp21dBx/sR27FAcovOumfOEoxVyvPMrjrhxo2vV50u90aNVTMuEEs33sMG0Hh3A9qEreQi9WC+vU1WTsAjk/jz16eBUo2w6jAZ1nRLpgEA8ZjqdEwxFaKdCwtnWaeDAwASkFQmvCX
            "text/plain": [
              "<Figure size 1100x800 with 3 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Display your beautiful creations!\n",
        "%matplotlib inline\n",
        "\n",
        "# figure size in inches optional\n",
        "rcParams[\"figure.figsize\"] = 11 ,8\n",
        "\n",
        "# read images\n",
        "img_A = mpimg.imread(edit_filepaths[0])\n",
        "img_B = mpimg.imread(edit_filepaths[1])\n",
        "img_C = mpimg.imread(edit_filepaths[2])\n",
        "\n",
        "# display images\n",
        "fig, ax = plt.subplots(1,3)\n",
        "[a.axis(\"off\") for a in ax]\n",
        "ax[0].imshow(img_A)\n",
        "ax[1].imshow(img_B)\n",
        "ax[2].imshow(img_C)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AK-BHQDMxUT0"
      },
      "source": [
        "Beautiful!\n",
        "\n",
        "Now you too can easily create dynamic masks to extend your images - enjoy the APIs, and please share what you build!"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}