MeshroomCollab.ipynb

@donmahallem I get the following error importing the ipynb:

There was an error loading this notebook. Ensure that the file is accessible and try again.
JSON.parse: unexpected character at line 1 column 1 of the JSON data

Copy paste to new project works.

Thanks for the headsup. Its fixed. Currently unsure what to add. As the workflows of providing the data can be quite different. Either mounting gdrive, upload zip or download from url.

Maybe an exampe on how to enable AKAZE in DescriberTypes on FeatureExtraction, FeatureMatching and StructureFromMotion?
https://github.com/alicevision/meshroom/wiki/Reconstruction-parameters

As I am fairly new to Meshroom there no direct CLI commands(atleast I couldnt find any), right? My currrent workflow would be creating a meshroom pipeline file on my current pc and upload it for processing to Colab.

I am working on the CLI documentation.
Basically you can get the available parameters by running the nodes in the cli without parameters or with --help.

I was looking into https://github.com/mikedh/trimesh - maybe there is a way to preview the 3d mesh in the notebook...

@donmahallem use "image dataset" instead of "meshfile"

Will do once not on a phone.

Regarding the visualization I took a quick glance at importing html into a notebook. Maybe this way one can integrate an interactive preview. Depends on vertices size/performance tho. Otherwise there are wavefront renderes available that could produce stills for 3-4 directions and embed those as images. Will look into it

Good idea, I think this will do: https://kitware.github.io/vtk-js-datasets/list/obj.html

Can someone please help?
Hi,
I get the exported .obj file as a point area cloud not as a mesh. I am opening it in blender. Thanks!

Here is my more refined notebook

My version (GDRIVE usage, bypassing the space issue)

{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"name": "meshroom_test.ipynb",
"version": "0.3.2",
"provenance": []
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
},
"accelerator": "GPU"
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "RD3eZSm6d6NZ",
"colab_type": "text"
},
"source": [
"#Setup\n",
"Download Meshroom and Data\n"
]
},
{
"cell_type": "code",
"metadata": {
"id": "cQgXdvk-qUvi",
"colab_type": "code",
"colab": {}
},
"source": [
"!git clone https://github.com/alicevision/dataset_buddha\n",
"!wget -N https://github.com/alicevision/meshroom/releases/download/v2019.1.0/Meshroom-2019.1.0-linux.tar.gz\n",
"!mkdir meshroom\n",
"!tar xzf Meshroom-2019.1.0-linux.tar.gz -C ./meshroom"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "zTI6AlUXd-_M",
"colab_type": "text"
},
"source": [
"#Upload data\n",
"Optional upload of pipline file and image source files\n"
]
},
{
"cell_type": "code",
"metadata": {
"id": "9_BfARNUdd_a",
"colab_type": "code",
"colab": {}
},
"source": [
"from google.colab import files\n",
"\n",
"# optional upload for the meshfile\n",
"\n",
"uploaded = files.upload()\n",
"\n",
"for fn in uploaded.keys():\n",
" print('User uploaded file "{name}" with length {length} bytes'.format( name=fn, length=len(uploaded[fn])))\n"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "xA_XKpKqdoqb",
"colab_type": "text"
},
"source": [
"#Meshing\n",
"use the --pipeline argument to provide a path to the meshfile you created/uploaded"
]
},
{
"cell_type": "code",
"metadata": {
"id": "AZWyk461sxGI",
"colab_type": "code",
"colab": {}
},
"source": [
"!mkdir ./buddha_out\n",
"!./meshroom/Meshroom-2019.1.0/meshroom_photogrammetry --input ./dataset_buddha/buddha --output ./buddha_out\n"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "7wwo3nPyd0-w",
"colab_type": "text"
},
"source": [
"#Download\n",
"Use the prefered download format\n"
]
},
{
"cell_type": "code",
"metadata": {
"id": "SPEdzNMo5fV6",
"colab_type": "code",
"colab": {}
},
"source": [
"!tar -czvf meshbuddha.tar.gz ./buddha_out\n",
"from google.colab import files\n",
"\n",
"\n",
"files.download('meshbuddha.tar.gz')"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "rAXWX6SX8J_q",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 85
},
"outputId": "168b3acd-bd30-47c7-ca77-b6870306dd78"
},
"source": [
"!zip -r meshbuddha.zip ./buddha_out\n",
"files.download('meshbuddha.zip')"
],
"execution_count": 19,
"outputs": [
{
"output_type": "stream",
"text": [
" adding: buddha_out/ (stored 0%)\n",
" adding: buddha_out/texture_0.png (deflated 0%)\n",
" adding: buddha_out/texturedMesh.obj (deflated 72%)\n",
" adding: buddha_out/texturedMesh.mtl (deflated 28%)\n"
],
"name": "stdout"
}
]
}
]
}

Hi!
Due to 2021.1.0 not being available I modified the file, to be more "actually", although there is still something to be done about few things. Some things could be resolved to give Meshroom some new features like file/folder remapings for different environments.
{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"Kopija beležnice RunMeshroom.ipynb","provenance":[{"file_id":"1qiHD3s4l_WCao0yIL6TIjvntY-Bgt32W","timestamp":1622313897316}],"collapsed_sections":[],"toc_visible":true},"kernelspec":{"name":"python3","display_name":"Python 3"},"accelerator":"GPU"},"cells":[{"cell_type":"markdown","metadata":{"id":"tJUlZN4TRnCH"},"source":["This notebook is based off of donmahallem's MeshroomColab notebook modified by shanji97.\n","\n","If you need help, check out his YouTube and his GitHub."]},{"cell_type":"markdown","metadata":{"id":"RD3eZSm6d6NZ"},"source":["#Setup (2021.1.0 version)\n","Download the meshroom tar file from Fosshub and upload it to a download service of you choice. Download the file and copy the direct download link (not the link the site is hosted on - It is usually formated as download.website.domain/someRandomIdString, here's an example: https://download.wetransfer.com//eugv/427e33c9565913f74fb2012ba98c895020210518183951/e837b9ffcec53fe59d42ac89c665f0134e3784bd/Meshroom-2021.1.0-linux-cuda10.tar.gz?cf.....). \n","\n","1) Set the download to the first input field. (TODO: Let the user choose between custom link and drive link.\n","\n","2) Set the path of your meshroom file\n","\n","3) Set the path of the (most recent) camera sensors database file (TODO: git clone and git pull the camerasensors.db file to Google Drive folder, manipulate the camera sensor location in the .mg file.\n","\n","4)(TODO): Remap the locations of the images using sed\n","\n"]},{"cell_type":"markdown","metadata":{"id":"PvIwyegUobtF"},"source":[""]},{"cell_type":"code","metadata":{"id":"Ol5uQcZkLvA3"},"source":["link_to_direct_download = 'https://download.wetransfer.com//eugv/427e33c9565913f74fb2012ba98c895020210518183951/e837b9ffcec53fe59d42ac89c665f0134e3784bd/Meshroom-2021.1.0-linux-cuda10.tar.gz?cf=y&token=eyJhbGciOiJIUzI1NiJ9.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.6iHqkKICcqXxZpMcWL_aHQ7BT6WfxJ_r9zVEjqvmHzo'#@param {type:"string"}\n","\n","meshroom_file_path = "/content/drive/MyDrive/Osnove_oblikovanja_2021/Projekt_1/MeshroomCaches/shanji/D2/BB.mg" #@param {type:"string"}\n","\n","sensorDB="/content/drive/MyDrive/Osnove_oblikovanja_2021/Projekt_1/MeshroomCaches/shanji/D2/cameraSensors.db" #@param {type:"string"}\n","\n","output_path="/content/drive/MyDrive/Osnove_oblikovanja_2021/Projekt_1/MeshroomCaches/shanji/D2/Cache" #@param {type:"string"}\n"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"kKyatDmQQsWA"},"source":[""],"execution_count":null,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"RsxzdzwtPVCA"},"source":["Create a directory to store the meshroom bins in and extract the meshroom bins there."]},{"cell_type":"code","metadata":{"id":"cQgXdvk-qUvi"},"source":["\n","\n","!mkdir ~/meshroom\n","!wget -N --user-agent Mozilla/4.0 -O ~/meshroom/MRoom.tar.gz '{link_to_direct_download}'\n","!tar -xvf ~/meshroom/MRoom.tar.gz\n"],"execution_count":null,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"zTI6AlUXd-_M"},"source":["#Google Drive\n","Connect to Google Drive, so that you can import and export files from there.\n"]},{"cell_type":"code","metadata":{"id":"2wodsIcpYbDR","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1621417172047,"user_tz":-120,"elapsed":9195,"user":{"displayName":"ALEKSANDER KOVA","photoUrl":"","userId":"13312320104478609314"}},"outputId":"6542250a-97ea-4b16-e024-51dc4981b05c"},"source":["from google.colab import drive\n","drive.mount('/content/drive')"],"execution_count":null,"outputs":[{"output_type":"stream","text":["Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount("/content/drive", force_remount=True).\n"],"name":"stdout"}]},{"cell_type":"markdown","metadata":{"id":"xA_XKpKqdoqb"},"source":["#Meshing\n","Use the form in setup to specify an input and output path, or check sample data to try it out."]},{"cell_type":"code","metadata":{"id":"AZWyk461sxGI"},"source":["!cd ~/meshroom\n","!Meshroom-2021.1.0-av2.4.0-centos7-cuda10.2/meshroom_compute --forceCompute '{meshroom_file_path}'"],"execution_count":null,"outputs":[]}]}