Skip to content

Instantly share code, notes, and snippets.

@armindocachada

armindocachada/pifu-demo-with-pytorch-downgraded.ipynb

Created August 6, 2022 16:20
Show Gist options
  • Save armindocachada/af1b21ef57d8f07516f36c76b20950c7 to your computer and use it in GitHub Desktop.
Save armindocachada/af1b21ef57d8f07516f36c76b20950c7 to your computer and use it in GitHub Desktop.
Download ZIP
PIFu Demo with pytorch downgraded
Display the source blob
Display the rendered blob
Raw
pifu-demo-with-pytorch-downgraded.ipynb
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"name": "PIFu Demo with pytorch downgraded",
"provenance": [],
"collapsed_sections": [],
"include_colab_link": true
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
},
"accelerator": "GPU"
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "view-in-github",
"colab_type": "text"
},
"source": [
"<a href=\"https://colab.research.google.com/gist/armindocachada/af1b21ef57d8f07516f36c76b20950c7/pifu-demo-with-pytorch-downgraded.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "LWTVk_PITxiE"
},
"source": [
"# PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "wmFdsTvLKtBO"
},
"source": [
"Demo of the original PyTorch based implementation provided here: https://github.com/shunsukesaito/PIFu\n",
"\n",
"## Note\n",
"Before running this notebook make sure that your runtime type is 'Python 3 with GPU acceleration'. Go to Edit > Notebook settings > Hardware Accelerator > Select \"GPU\".\n",
"\n",
"## More Info\n",
"- Paper: https://arxiv.org/pdf/1905.05172.pdf\n",
"- Repo: https://github.com/shunsukesaito/PIFu\n",
"- Project Page: https://shunsukesaito.github.io/PIFu/"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "8vZaAyhUJ9QC"
},
"source": [
"## Requirements\n",
"- Python 3\n",
"- PyTorch tested on 1.4.0\n",
"- json\n",
"- PIL\n",
"- skimage\n",
"- tqdm\n",
"- numpy\n",
"- cv2"
]
},
{
"cell_type": "code",
"metadata": {
"id": "M9Az9CQpTT8_",
"colab": {
"base_uri": "https://localhost:8080/"
},
"outputId": "86cd915d-ecdf-42ac-af8a-aefc6ad4ac03"
},
"source": [
"!git clone https://github.com/shunsukesaito/PIFu.git"
],
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Cloning into 'PIFu'...\n",
"remote: Enumerating objects: 218, done.\u001b[K\n",
"remote: Counting objects: 100% (11/11), done.\u001b[K\n",
"remote: Compressing objects: 100% (10/10), done.\u001b[K\n",
"remote: Total 218 (delta 4), reused 6 (delta 1), pack-reused 207\u001b[K\n",
"Receiving objects: 100% (218/218), 2.43 MiB | 22.62 MiB/s, done.\n",
"Resolving deltas: 100% (90/90), done.\n"
]
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "Y0rgMInwTt0s"
},
"source": [
"## Demo"
]
},
{
"cell_type": "code",
"metadata": {
"id": "UrIcZweSNRFI",
"colab": {
"base_uri": "https://localhost:8080/"
},
"outputId": "45b4b517-925d-4773-dfa4-89bd4ec5cf84"
},
"source": [
"cd /content/PIFu"
],
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"/content/PIFu\n"
]
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "k3jjm6HuQRk8",
"colab": {
"base_uri": "https://localhost:8080/"
},
"outputId": "062c251e-61e0-49e9-d1b3-3ad4f051f665"
},
"source": [
"!sh ./scripts/download_trained_model.sh"
],
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"+ mkdir -p checkpoints\n",
"+ cd checkpoints\n",
"+ wget https://drive.google.com/uc?export=download&id=1zEmVXG2VHy0MMzngcRshB4D8Sr_oLHsm -O net_G\n",
"--2022-08-06 15:35:02-- https://drive.google.com/uc?export=download&id=1zEmVXG2VHy0MMzngcRshB4D8Sr_oLHsm\n",
"Resolving drive.google.com (drive.google.com)... 74.125.196.113, 74.125.196.102, 74.125.196.101, ...\n",
"Connecting to drive.google.com (drive.google.com)|74.125.196.113|:443... connected.\n",
"HTTP request sent, awaiting response... 303 See Other\n",
"Location: https://doc-14-0o-docs.googleusercontent.com/docs/securesc/ha0ro937gcuc7l7deffksulhg5h7mbp1/u66c4r86q1p8auq46lfk2dipmatj3jgb/1659800100000/14428417455748421393/*/1zEmVXG2VHy0MMzngcRshB4D8Sr_oLHsm?e=download&uuid=10de986b-43fb-45a1-b26f-088ac7cc7891 [following]\n",
"Warning: wildcards not supported in HTTP.\n",
"--2022-08-06 15:35:07-- https://doc-14-0o-docs.googleusercontent.com/docs/securesc/ha0ro937gcuc7l7deffksulhg5h7mbp1/u66c4r86q1p8auq46lfk2dipmatj3jgb/1659800100000/14428417455748421393/*/1zEmVXG2VHy0MMzngcRshB4D8Sr_oLHsm?e=download&uuid=10de986b-43fb-45a1-b26f-088ac7cc7891\n",
"Resolving doc-14-0o-docs.googleusercontent.com (doc-14-0o-docs.googleusercontent.com)... 172.217.193.132, 2607:f8b0:400c:c03::84\n",
"Connecting to doc-14-0o-docs.googleusercontent.com (doc-14-0o-docs.googleusercontent.com)|172.217.193.132|:443... connected.\n",
"HTTP request sent, awaiting response... 200 OK\n",
"Length: 62510909 (60M) [application/octet-stream]\n",
"Saving to: ‘net_G’\n",
"\n",
"net_G 100%[===================>] 59.61M 336MB/s in 0.2s \n",
"\n",
"2022-08-06 15:35:07 (336 MB/s) - ‘net_G’ saved [62510909/62510909]\n",
"\n",
"+ wget https://drive.google.com/uc?export=download&id=1V83B6GDIjYMfHdpg-KcCSAPgHxpafHgd -O net_C\n",
"--2022-08-06 15:35:07-- https://drive.google.com/uc?export=download&id=1V83B6GDIjYMfHdpg-KcCSAPgHxpafHgd\n",
"Resolving drive.google.com (drive.google.com)... 74.125.196.113, 74.125.196.102, 74.125.196.101, ...\n",
"Connecting to drive.google.com (drive.google.com)|74.125.196.113|:443... connected.\n",
"HTTP request sent, awaiting response... 303 See Other\n",
"Location: https://doc-00-0o-docs.googleusercontent.com/docs/securesc/ha0ro937gcuc7l7deffksulhg5h7mbp1/ee4fa845p3arp1ct878if9hu9qoppn43/1659800100000/14428417455748421393/*/1V83B6GDIjYMfHdpg-KcCSAPgHxpafHgd?e=download&uuid=994d6160-ac54-490f-bdf0-c45900bb3d77 [following]\n",
"Warning: wildcards not supported in HTTP.\n",
"--2022-08-06 15:35:08-- https://doc-00-0o-docs.googleusercontent.com/docs/securesc/ha0ro937gcuc7l7deffksulhg5h7mbp1/ee4fa845p3arp1ct878if9hu9qoppn43/1659800100000/14428417455748421393/*/1V83B6GDIjYMfHdpg-KcCSAPgHxpafHgd?e=download&uuid=994d6160-ac54-490f-bdf0-c45900bb3d77\n",
"Resolving doc-00-0o-docs.googleusercontent.com (doc-00-0o-docs.googleusercontent.com)... 172.217.193.132, 2607:f8b0:400c:c03::84\n",
"Connecting to doc-00-0o-docs.googleusercontent.com (doc-00-0o-docs.googleusercontent.com)|172.217.193.132|:443... connected.\n",
"HTTP request sent, awaiting response... 200 OK\n",
"Length: 36570643 (35M) [application/octet-stream]\n",
"Saving to: ‘net_C’\n",
"\n",
"net_C 100%[===================>] 34.88M 169MB/s in 0.2s \n",
"\n",
"2022-08-06 15:35:09 (169 MB/s) - ‘net_C’ saved [36570643/36570643]\n",
"\n",
"+ cd ..\n"
]
}
]
},
{
"cell_type": "code",
"source": [
"!pip install torch==1.4.0"
],
"metadata": {
"id": "fr2fAD4h4ktw",
"outputId": "c89d37ff-cc00-4eb8-a40e-01c1a42d3320",
"colab": {
"base_uri": "https://localhost:8080/"
}
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
"Collecting torch==1.4.0\n",
" Downloading torch-1.4.0-cp37-cp37m-manylinux1_x86_64.whl (753.4 MB)\n",
"\u001b[K |████████████████████████████████| 753.4 MB 7.1 kB/s \n",
"\u001b[?25hInstalling collected packages: torch\n",
" Attempting uninstall: torch\n",
" Found existing installation: torch 1.12.0+cu113\n",
" Uninstalling torch-1.12.0+cu113:\n",
" Successfully uninstalled torch-1.12.0+cu113\n",
"\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
"torchvision 0.13.0+cu113 requires torch==1.12.0, but you have torch 1.4.0 which is incompatible.\n",
"torchtext 0.13.0 requires torch==1.12.0, but you have torch 1.4.0 which is incompatible.\n",
"torchaudio 0.12.0+cu113 requires torch==1.12.0, but you have torch 1.4.0 which is incompatible.\n",
"fastai 2.7.7 requires torch<1.13,>=1.7, but you have torch 1.4.0 which is incompatible.\u001b[0m\n",
"Successfully installed torch-1.4.0\n"
]
}
]
},
{
"cell_type": "code",
"source": [
"!pip install torchvision==0.5.0"
],
"metadata": {
"id": "YchK6vyd6Bls",
"outputId": "f8f5ac54-da37-44d1-9aa8-1cf61a4a17c0",
"colab": {
"base_uri": "https://localhost:8080/"
}
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
"Collecting torchvision==0.5.0\n",
" Downloading torchvision-0.5.0-cp37-cp37m-manylinux1_x86_64.whl (4.0 MB)\n",
"\u001b[K |████████████████████████████████| 4.0 MB 3.2 MB/s \n",
"\u001b[?25hRequirement already satisfied: pillow>=4.1.1 in /usr/local/lib/python3.7/dist-packages (from torchvision==0.5.0) (7.1.2)\n",
"Requirement already satisfied: numpy in /usr/local/lib/python3.7/dist-packages (from torchvision==0.5.0) (1.21.6)\n",
"Requirement already satisfied: torch==1.4.0 in /usr/local/lib/python3.7/dist-packages (from torchvision==0.5.0) (1.4.0)\n",
"Requirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from torchvision==0.5.0) (1.15.0)\n",
"Installing collected packages: torchvision\n",
" Attempting uninstall: torchvision\n",
" Found existing installation: torchvision 0.13.0+cu113\n",
" Uninstalling torchvision-0.13.0+cu113:\n",
" Successfully uninstalled torchvision-0.13.0+cu113\n",
"\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
"fastai 2.7.7 requires torch<1.13,>=1.7, but you have torch 1.4.0 which is incompatible.\n",
"fastai 2.7.7 requires torchvision>=0.8.2, but you have torchvision 0.5.0 which is incompatible.\u001b[0m\n",
"Successfully installed torchvision-0.5.0\n"
]
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "6heKcA-0QEBw"
},
"source": [
"# RUN PIFu\n"
]
},
{
"cell_type": "code",
"source": [
"import torch;\n",
"torch.__version__"
],
"metadata": {
"id": "fgwl0z-p23rn",
"outputId": "548d69b8-7be4-43b5-9849-8045a8b4ede7",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 35
}
},
"execution_count": null,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"'1.4.0'"
],
"application/vnd.google.colaboratory.intrinsic+json": {
"type": "string"
}
},
"metadata": {},
"execution_count": 6
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "5995t2PnQTmG",
"colab": {
"base_uri": "https://localhost:8080/"
},
"outputId": "24ffefca-b4f2-4d37-84c2-076d090ee482"
},
"source": [
"!sh ./scripts/test.sh"
],
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"+ GPU_ID=0\n",
"+ DISPLAY_ID=10\n",
"+ NAME=pifu_demo\n",
"+ BATCH_SIZE=1\n",
"+ MLP_DIM=257 1024 512 256 128 1\n",
"+ MLP_DIM_COLOR=513 1024 512 256 128 3\n",
"+ VOL_RES=256\n",
"+ CHECKPOINTS_NETG_PATH=./checkpoints/net_G\n",
"+ CHECKPOINTS_NETC_PATH=./checkpoints/net_C\n",
"+ TEST_FOLDER_PATH=./sample_images\n",
"+ CUDA_VISIBLE_DEVICES=0 python ./apps/eval.py --name pifu_demo --batch_size 1 --mlp_dim 257 1024 512 256 128 1 --mlp_dim_color 513 1024 512 256 128 3 --num_stack 4 --num_hourglass 2 --resolution 256 --hg_down ave_pool --norm group --norm_color group --test_folder_path ./sample_images --load_netG_checkpoint_path ./checkpoints/net_G --load_netC_checkpoint_path ./checkpoints/net_C\n",
"initialize network with normal\n",
"Using Network: hgpifu\n",
"loading for net C ... ./checkpoints/net_C\n",
"initialize network with normal\n",
"num; 1\n",
"0it [00:00, ?it/s]./sample_images/codemental_2.png ./sample_images/codemental_2_mask.png\n",
"/content/PIFu/lib/mesh_util.py:45: FutureWarning: marching_cubes_lewiner is deprecated in favor of marching_cubes. marching_cubes_lewiner will be removed in version 0.19\n",
" verts, faces, normals, values = measure.marching_cubes_lewiner(sdf, 0.5)\n",
"1it [00:06, 6.57s/it]\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
"# Training Demo\n",
"\n",
"Currently it is not possible to run the training in Google Colab. Best option is to run it in Linux or using a Dockerfile (see my blog article at spltech.co.uk/blog)"
],
"metadata": {
"id": "LyJzIKhQMOQb"
}
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment