{ "cells": [ { "cell_type": "markdown", "metadata": { "id": "N6ZDpd9XzFeN" }, "source": [ "##### Copyright 2018 The TensorFlow Hub Authors.\n", "\n", "Licensed under the Apache License, Version 2.0 (the \"License\");" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "cellView": "both", "execution": { "iopub.execute_input": "2023-05-12T12:32:28.887784Z", "iopub.status.busy": "2023-05-12T12:32:28.887334Z", "iopub.status.idle": "2023-05-12T12:32:28.891355Z", "shell.execute_reply": "2023-05-12T12:32:28.890747Z" }, "id": "KUu4vOt5zI9d" }, "outputs": [], "source": [ "# Copyright 2018 The TensorFlow Hub Authors. All Rights Reserved.\n", "#\n", "# Licensed under the Apache License, Version 2.0 (the \"License\");\n", "# you may not use this file except in compliance with the License.\n", "# You may obtain a copy of the License at\n", "#\n", "# http://www.apache.org/licenses/LICENSE-2.0\n", "#\n", "# Unless required by applicable law or agreed to in writing, software\n", "# distributed under the License is distributed on an \"AS IS\" BASIS,\n", "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n", "# See the License for the specific language governing permissions and\n", "# limitations under the License.\n", "# ==============================================================================" ] }, { "cell_type": "markdown", "metadata": { "id": "CxmDMK4yupqg" }, "source": [ "# Generate Artificial Faces with CelebA Progressive GAN Model\n" ] }, { "cell_type": "markdown", "metadata": { "id": "MfBg1C5NB3X0" }, "source": [ "\n", " \n", " \n", " \n", " \n", " \n", "
\n", " View on TensorFlow.org\n", " \n", " Run in Google Colab\n", " \n", " View on GitHub\n", " \n", " Download notebook\n", " \n", " See TF Hub model\n", "
" ] }, { "cell_type": "markdown", "metadata": { "id": "Sy553YSVmYiK" }, "source": [ "This Colab demonstrates use of a TF Hub module based on a generative adversarial network (GAN). The module maps from N-dimensional vectors, called latent space, to RGB images.\n", "\n", "Two examples are provided:\n", "* **Mapping** from latent space to images, and\n", "* Given a target image, **using gradient descent to find** a latent vector that generates an image similar to the target image." ] }, { "cell_type": "markdown", "metadata": { "id": "v4XGxDrCkeip" }, "source": [ "## Optional prerequisites\n", "\n", "* Familiarity with [low level Tensorflow concepts](https://www.tensorflow.org/guide/eager).\n", "* [Generative Adversarial Network](https://en.wikipedia.org/wiki/Generative_adversarial_network) on Wikipedia.\n", "* Paper on Progressive GANs: [Progressive Growing of GANs for Improved Quality, Stability, and Variation](https://arxiv.org/abs/1710.10196)." ] }, { "cell_type": "markdown", "metadata": { "id": "HK3Q2vIaVw56" }, "source": [ "### More models\n", "[Here](https://tfhub.dev/s?module-type=image-generator) you can find all models currently hosted on [tfhub.dev](https://tfhub.dev/) that can generate images." ] }, { "cell_type": "markdown", "metadata": { "id": "Q4DN769E2O_R" }, "source": [ "## Setup" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "execution": { "iopub.execute_input": "2023-05-12T12:32:28.895311Z", "iopub.status.busy": "2023-05-12T12:32:28.895092Z", "iopub.status.idle": "2023-05-12T12:32:37.573232Z", "shell.execute_reply": "2023-05-12T12:32:37.572427Z" }, "id": "KNM3kA0arrUu" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Collecting git+https://github.com/tensorflow/docs\r\n", " Cloning https://github.com/tensorflow/docs to /tmpfs/tmp/pip-req-build-qiu5de7c\r\n", " Running command git clone --filter=blob:none --quiet https://github.com/tensorflow/docs /tmpfs/tmp/pip-req-build-qiu5de7c\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Resolved https://github.com/tensorflow/docs to commit 5d7c4c291249a7c7e149316ef0ac5eb86fd2cda1\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Preparing metadata (setup.py) ... \u001b[?25l-" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b \bdone\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\u001b[?25hCollecting astor (from tensorflow-docs==0.0.0.dev0)\r\n", " Using cached astor-0.8.1-py2.py3-none-any.whl (27 kB)\r\n", "Requirement already satisfied: absl-py in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow-docs==0.0.0.dev0) (1.4.0)\r\n", "Requirement already satisfied: jinja2 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow-docs==0.0.0.dev0) (3.1.2)\r\n", "Requirement already satisfied: nbformat in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow-docs==0.0.0.dev0) (5.8.0)\r\n", "Requirement already satisfied: protobuf>=3.12 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow-docs==0.0.0.dev0) (4.23.0)\r\n", "Requirement already satisfied: pyyaml in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow-docs==0.0.0.dev0) (6.0)\r\n", "Requirement already satisfied: MarkupSafe>=2.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from jinja2->tensorflow-docs==0.0.0.dev0) (2.1.2)\r\n", "Requirement already satisfied: fastjsonschema in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from nbformat->tensorflow-docs==0.0.0.dev0) (2.16.3)\r\n", "Requirement already satisfied: jsonschema>=2.6 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from nbformat->tensorflow-docs==0.0.0.dev0) (4.17.3)\r\n", "Requirement already satisfied: jupyter-core in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from nbformat->tensorflow-docs==0.0.0.dev0) (5.3.0)\r\n", "Requirement already satisfied: traitlets>=5.1 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from nbformat->tensorflow-docs==0.0.0.dev0) (5.9.0)\r\n", "Requirement already satisfied: attrs>=17.4.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from jsonschema>=2.6->nbformat->tensorflow-docs==0.0.0.dev0) (23.1.0)\r\n", "Requirement already satisfied: pyrsistent!=0.17.0,!=0.17.1,!=0.17.2,>=0.14.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from jsonschema>=2.6->nbformat->tensorflow-docs==0.0.0.dev0) (0.19.3)\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Requirement already satisfied: platformdirs>=2.5 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from jupyter-core->nbformat->tensorflow-docs==0.0.0.dev0) (3.5.1)\r\n", "Building wheels for collected packages: tensorflow-docs\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Building wheel for tensorflow-docs (setup.py) ... \u001b[?25l-" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b \b\\" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b \bdone\r\n", "\u001b[?25h Created wheel for tensorflow-docs: filename=tensorflow_docs-0.0.0.dev0-py3-none-any.whl size=183463 sha256=ddabb4b4e55b99da2be9992923caec7a6d75998699e2d2e27f0e6ee80c10a2c3\r\n", " Stored in directory: /tmpfs/tmp/pip-ephem-wheel-cache-l38io7za/wheels/fc/f8/3b/5d21409a59cb1be9b1ade11f682039ced75b84de9dd6a0c8de\r\n", "Successfully built tensorflow-docs\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Installing collected packages: astor, tensorflow-docs\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Successfully installed astor-0.8.1 tensorflow-docs-0.0.0.dev0\r\n" ] } ], "source": [ "# Install imageio for creating animations. \n", "!pip -q install imageio\n", "!pip -q install scikit-image\n", "!pip install git+https://github.com/tensorflow/docs" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "cellView": "both", "execution": { "iopub.execute_input": "2023-05-12T12:32:37.577299Z", "iopub.status.busy": "2023-05-12T12:32:37.577046Z", "iopub.status.idle": "2023-05-12T12:32:40.326382Z", "shell.execute_reply": "2023-05-12T12:32:40.325666Z" }, "id": "6cPY9Ou4sWs_" }, "outputs": [], "source": [ "#@title Imports and function definitions\n", "from absl import logging\n", "\n", "import imageio\n", "import PIL.Image\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", "\n", "import tensorflow as tf\n", "tf.random.set_seed(0)\n", "\n", "import tensorflow_hub as hub\n", "from tensorflow_docs.vis import embed\n", "import time\n", "\n", "try:\n", " from google.colab import files\n", "except ImportError:\n", " pass\n", "\n", "from IPython import display\n", "from skimage import transform\n", "\n", "# We could retrieve this value from module.get_input_shapes() if we didn't know\n", "# beforehand which module we will be using.\n", "latent_dim = 512\n", "\n", "\n", "# Interpolates between two vectors that are non-zero and don't both lie on a\n", "# line going through origin. First normalizes v2 to have the same norm as v1. \n", "# Then interpolates between the two vectors on the hypersphere.\n", "def interpolate_hypersphere(v1, v2, num_steps):\n", " v1_norm = tf.norm(v1)\n", " v2_norm = tf.norm(v2)\n", " v2_normalized = v2 * (v1_norm / v2_norm)\n", "\n", " vectors = []\n", " for step in range(num_steps):\n", " interpolated = v1 + (v2_normalized - v1) * step / (num_steps - 1)\n", " interpolated_norm = tf.norm(interpolated)\n", " interpolated_normalized = interpolated * (v1_norm / interpolated_norm)\n", " vectors.append(interpolated_normalized)\n", " return tf.stack(vectors)\n", "\n", "# Simple way to display an image.\n", "def display_image(image):\n", " image = tf.constant(image)\n", " image = tf.image.convert_image_dtype(image, tf.uint8)\n", " return PIL.Image.fromarray(image.numpy())\n", "\n", "# Given a set of images, show an animation.\n", "def animate(images):\n", " images = np.array(images)\n", " converted_images = np.clip(images * 255, 0, 255).astype(np.uint8)\n", " imageio.mimsave('./animation.gif', converted_images)\n", " return embed.embed_file('./animation.gif')\n", "\n", "logging.set_verbosity(logging.ERROR)" ] }, { "cell_type": "markdown", "metadata": { "id": "f5EESfBvukYI" }, "source": [ "## Latent space interpolation" ] }, { "cell_type": "markdown", "metadata": { "id": "nJb9gFmRvynZ" }, "source": [ "### Random vectors\n", "\n", "Latent space interpolation between two randomly initialized vectors. We will use a TF Hub module [progan-128](https://tfhub.dev/google/progan-128/1) that contains a pre-trained Progressive GAN." ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "execution": { "iopub.execute_input": "2023-05-12T12:32:40.330952Z", "iopub.status.busy": "2023-05-12T12:32:40.330304Z", "iopub.status.idle": "2023-05-12T12:32:45.400081Z", "shell.execute_reply": "2023-05-12T12:32:45.399319Z" }, "id": "8StEe9x9wGma" }, "outputs": [], "source": [ "progan = hub.load(\"https://tfhub.dev/google/progan-128/1\").signatures['default']" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "execution": { "iopub.execute_input": "2023-05-12T12:32:45.404218Z", "iopub.status.busy": "2023-05-12T12:32:45.403675Z", "iopub.status.idle": "2023-05-12T12:32:53.096661Z", "shell.execute_reply": "2023-05-12T12:32:53.096038Z" }, "id": "fZ0O5_5Jhwio" }, "outputs": [ { "data": { "text/html": [ "" ], "text/plain": [ "" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "def interpolate_between_vectors():\n", " v1 = tf.random.normal([latent_dim])\n", " v2 = tf.random.normal([latent_dim])\n", " \n", " # Creates a tensor with 25 steps of interpolation between v1 and v2.\n", " vectors = interpolate_hypersphere(v1, v2, 50)\n", "\n", " # Uses module to generate images from the latent space.\n", " interpolated_images = progan(vectors)['default']\n", "\n", " return interpolated_images\n", "\n", "interpolated_images = interpolate_between_vectors()\n", "animate(interpolated_images)" ] }, { "cell_type": "markdown", "metadata": { "id": "L9-uXoTHuXQC" }, "source": [ "## Finding closest vector in latent space\n", "Fix a target image. As an example use an image generated from the module or upload your own." ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "cellView": "both", "execution": { "iopub.execute_input": "2023-05-12T12:32:53.107195Z", "iopub.status.busy": "2023-05-12T12:32:53.106532Z", "iopub.status.idle": "2023-05-12T12:32:53.493162Z", "shell.execute_reply": "2023-05-12T12:32:53.492557Z" }, "id": "phT4W66pMmko" }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "image_from_module_space = True # @param { isTemplate:true, type:\"boolean\" }\n", "\n", "def get_module_space_image():\n", " vector = tf.random.normal([1, latent_dim])\n", " images = progan(vector)['default'][0]\n", " return images\n", "\n", "def upload_image():\n", " uploaded = files.upload()\n", " image = imageio.imread(uploaded[list(uploaded.keys())[0]])\n", " return transform.resize(image, [128, 128])\n", "\n", "if image_from_module_space:\n", " target_image = get_module_space_image()\n", "else:\n", " target_image = upload_image()\n", "\n", "display_image(target_image)" ] }, { "cell_type": "markdown", "metadata": { "id": "rBIt3Q4qvhuq" }, "source": [ "After defining a loss function between the target image and the image generated by a latent space variable, we can use gradient descent to find variable values that minimize the loss." ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "execution": { "iopub.execute_input": "2023-05-12T12:32:53.496483Z", "iopub.status.busy": "2023-05-12T12:32:53.496255Z", "iopub.status.idle": "2023-05-12T12:32:53.505562Z", "shell.execute_reply": "2023-05-12T12:32:53.505001Z" }, "id": "cUGakLdbML2Q" }, "outputs": [], "source": [ "tf.random.set_seed(42)\n", "initial_vector = tf.random.normal([1, latent_dim])" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "execution": { "iopub.execute_input": "2023-05-12T12:32:53.508545Z", "iopub.status.busy": "2023-05-12T12:32:53.508029Z", "iopub.status.idle": "2023-05-12T12:32:53.604739Z", "shell.execute_reply": "2023-05-12T12:32:53.604177Z" }, "id": "u7MGzDE5MU20" }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "display_image(progan(initial_vector)['default'][0])" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "execution": { "iopub.execute_input": "2023-05-12T12:32:53.607915Z", "iopub.status.busy": "2023-05-12T12:32:53.607406Z", "iopub.status.idle": "2023-05-12T12:33:01.172515Z", "shell.execute_reply": "2023-05-12T12:33:01.171831Z" }, "id": "q_4Z7tnyg-ZY" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "." ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "...\n", "....." ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "........" ] }, { "name": "stdout", "output_type": "stream", "text": [ "......." ] } ], "source": [ "def find_closest_latent_vector(initial_vector, num_optimization_steps,\n", " steps_per_image):\n", " images = []\n", " losses = []\n", "\n", " vector = tf.Variable(initial_vector) \n", " optimizer = tf.optimizers.Adam(learning_rate=0.01)\n", " loss_fn = tf.losses.MeanAbsoluteError(reduction=\"sum\")\n", "\n", " for step in range(num_optimization_steps):\n", " if (step % 100)==0:\n", " print()\n", " print('.', end='')\n", " with tf.GradientTape() as tape:\n", " image = progan(vector.read_value())['default'][0]\n", " if (step % steps_per_image) == 0:\n", " images.append(image.numpy())\n", " target_image_difference = loss_fn(image, target_image[:,:,:3])\n", " # The latent vectors were sampled from a normal distribution. We can get\n", " # more realistic images if we regularize the length of the latent vector to \n", " # the average length of vector from this distribution.\n", " regularizer = tf.abs(tf.norm(vector) - np.sqrt(latent_dim))\n", " \n", " loss = target_image_difference + regularizer\n", " losses.append(loss.numpy())\n", " grads = tape.gradient(loss, [vector])\n", " optimizer.apply_gradients(zip(grads, [vector]))\n", " \n", " return images, losses\n", "\n", "\n", "num_optimization_steps=200\n", "steps_per_image=5\n", "images, loss = find_closest_latent_vector(initial_vector, num_optimization_steps, steps_per_image)" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "execution": { "iopub.execute_input": "2023-05-12T12:33:01.175975Z", "iopub.status.busy": "2023-05-12T12:33:01.175557Z", "iopub.status.idle": "2023-05-12T12:33:01.342670Z", "shell.execute_reply": "2023-05-12T12:33:01.341791Z" }, "id": "pRbeF2oSAcOB" }, "outputs": [ { "data": { "text/plain": [ "(0.0, 6696.193258666992)" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plt.plot(loss)\n", "plt.ylim([0,max(plt.ylim())])" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "execution": { "iopub.execute_input": "2023-05-12T12:33:01.346267Z", "iopub.status.busy": "2023-05-12T12:33:01.345695Z", "iopub.status.idle": "2023-05-12T12:33:02.064630Z", "shell.execute_reply": "2023-05-12T12:33:02.063981Z" }, "id": "KnZkDy2FEsTt" }, "outputs": [ { "data": { "text/html": [ "" ], "text/plain": [ "" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "animate(np.stack(images))" ] }, { "cell_type": "markdown", "metadata": { "id": "GGKfuCdfPQKH" }, "source": [ "Compare the result to the target:" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "execution": { "iopub.execute_input": "2023-05-12T12:33:02.073675Z", "iopub.status.busy": "2023-05-12T12:33:02.073431Z", "iopub.status.idle": "2023-05-12T12:33:02.091151Z", "shell.execute_reply": "2023-05-12T12:33:02.090575Z" }, "id": "TK1P5z3bNuIl" }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "display_image(np.concatenate([images[-1], target_image], axis=1))" ] }, { "cell_type": "markdown", "metadata": { "id": "tDt15dLsJwMy" }, "source": [ "### Playing with the above example\n", "If image is from the module space, the descent is quick and converges to a reasonable sample. Try out descending to an image that is **not from the module space**. The descent will only converge if the image is reasonably close to the space of training images.\n", "\n", "How to make it descend faster and to a more realistic image? One can try:\n", "* using different loss on the image difference, e.g. quadratic,\n", "* using different regularizer on the latent vector,\n", "* initializing from a random vector in multiple runs,\n", "* etc.\n" ] } ], "metadata": { "accelerator": "GPU", "colab": { "collapsed_sections": [ "N6ZDpd9XzFeN" ], "name": "TF-Hub generative image module", "private_outputs": true, "provenance": [], "toc_visible": true }, "kernelspec": { "display_name": "Python 3", "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": "3.9.16" } }, "nbformat": 4, "nbformat_minor": 0 }