{ "info": { "author": "NVIDIA Corporation", "author_email": "info@nvidia.com", "bugtrack_url": null, "classifiers": [ "Development Status :: 5 - Production/Stable", "Intended Audience :: Developers", "Natural Language :: English", "Operating System :: OS Independent", "Programming Language :: Python :: 3.6", "Topic :: Software Development :: Libraries :: Python Modules", "Topic :: Utilities" ], "description": "# Nvidia Dataset Utilities (NVDU)\nThis project is a collection of Python scripts to help work with datasets for deep learning. For example, visualizing annotation data associated with captured sensor images generated by NVIDIA Deep learning Dataset Synthesizer (NDDS) https://github.com/NVIDIA/Dataset_Synthesizer.\n> **This module depends on OpenCV-python which currently doesn't work with Python 3.7.**\n\n![](./NVDUIntro.png)\n\n*Example of a dataset frame visualized using NVDU, showing axes and 3D cuboids for annotated objects.*\n\n## Table of Contents\n- [Nvidia Dataset Utilities (NVDU)](#nvidia-dataset-utilities-nvdu)\n - [Table of Contents](#table-of-contents)\n- [Install](#install)\n - [Install from pip:](#install-from-pip)\n - [Install from source code git repo:](#install-from-source-code-git-repo)\n- [nvdu_ycb](#nvdu_ycb)\n - [Usage](#usage)\n- [nvdu_viz](#nvdu_viz)\n - [Usage](#usage-1)\n - [Examples](#examples)\n - [Visualize a dataset generated by NDDS:](#visualize-a-dataset-generated-by-ndds)\n - [Visualize a set of images using different annotation data:](#visualize-a-set-of-images-using-different-annotation-data)\n - [Controls](#controls)\n - [Visualization options:](#visualization-options)\n - [Other:](#other)\n\n# Install\n## Install from pip:\n`pip install nvdu`\n\n## Install from source code git repo:\n**Clone the repo**\n\n_Using ssh path:_\n```\ngit clone ssh://git@github.com:12051/NVIDIA/Dataset_Utilities.git\n```\n_Using https path:_\n```\ngit clone https://github.com/NVIDIA/Dataset_Utilities.git\n```\n**Go inside the cloned repo's directory**\n```\ncd Dataset_Utilities\n```\n\n**Install locally**\n\n`pip install -e .`\n\n# nvdu_ycb\n_nvdu_ycb_ command help download, extract and align the YCB 3d models (which are used in the FAT dataset: http://research.nvidia.com/publication/2018-06_Falling-Things).\n## Usage\n```\nusage: nvdu_ycb [-h] [-s] [-l] [ycb_object_name]\n\nNVDU YCB models Support\n\npositional arguments:\n ycb_object_name Name of the YCB object to check.\n\noptional arguments:\n -h, --help show this help message and exit\n -s, --setup Setup the YCB models for the FAT dataset\n -l, --list List all the supported YCB objects\n```\n\n*NOTE: If you don't run the `nvdu_ycb --setup` before trying to use nvdu_viz, the visualizer will not be able to find the 3d models of the YCB object to overlay.*\n\n# nvdu_viz\n_nvdu_viz_ command visualizes the annotated datasets using the NDDS format.\n## Usage\n```\nnvdu_viz [-h] [-a DATA_ANNOT_DIR] [-s SIZE SIZE]\n [-o OBJECT_SETTINGS_PATH] [-c CAMERA_SETTINGS_PATH]\n [-m MODEL_DIR] [-n [NAME_FILTERS [NAME_FILTERS ...]]]\n [--fps FPS] [--auto_change] [-e EXPORT_DIR] [--auto_export]\n [--ignore_fixed_transform]\n [dataset_dir]\n\nNVDU Data Visualiser\n\npositional arguments:\n dataset_dir Dataset directory. This is where all the images\n (required) and annotation info (optional) are.\n Defaults to the current directory.\n\noptional arguments:\n -h, --help show this help message and exit.\n -a DATA_ANNOT_DIR, --data_annot_dir DATA_ANNOT_DIR\n Directory path - where to find the annotation data.\n Defaults to be the same directory as the dataset\n directory.\n -s SIZE SIZE, --size SIZE SIZE\n Window's size: [width, height]. If not specified then\n the window is sized to fit the resolution of the camera.\n -o OBJECT_SETTINGS_PATH, --object_settings_path OBJECT_SETTINGS_PATH\n Object settings file path.\n -c CAMERA_SETTINGS_PATH, --camera_settings_path CAMERA_SETTINGS_PATH\n Camera settings file path.\n -n [NAME_FILTERS [NAME_FILTERS ...]], --name_filters [NAME_FILTERS [NAME_FILTERS ...]]\n The name filter of each frame. e.g: *.png.\n --fps FPS How fast to automatically change frame.\n --auto_change When using this flag, the visualizer will automatically\n change the frame.\n -e EXPORT_DIR, --export_dir EXPORT_DIR\n Directory path - where to store the visualized images.\n If this is set, the script will automatically export\n the visualized image to the export directory.\n --auto_export When using this flag, the visualizer will automatically\n export the visualized frame to an image file in the\n `export_dir` directory.\n --ignore_fixed_transform\n When using this flag, the visualizer will not use the\n fixed transform matrix for the 3d model.\n```\n_NOTE: The `nvdu_viz` script can work from any directory_\n\n## Examples\n### Visualize a dataset generated by NDDS:\n1. Visualize the current directory:\n```\nnvdu_viz\n```\n2. Visualize a relative path:\n```\nnvdu_viz ../a_dataset\n```\n3. Visualize an absolute path:\n```\nnvdu_viz ~/data/dataset\n```\n4. Visualize different aspect of a frame using a filter:\n```\nnvdu_viz dataset_path --name_filters *.left.png *.right.png\n```\n\n### Visualize a set of images using different annotation data:\n1. The camera and object settings files are in the image directory:\n```\nnvdu_viz image_directory_here -a annotation_directory_here\n```\n2. The camera and object settings files are NOT in the image directory:\n```\nnvdu_viz image_directory_here -a annotation_directory_here -c camera_setting_path_here -o object_setting_path_here\n```\n\n## Controls\n### Visualization options:\n```\nF3 - Toggle the 2d cuboid\nF4 - Toggle the 3d cuboid\nF5 - Toggle the 3d models\nF6 - Toggle the axes\nF7 - Toggle the overlay frame name\n1 - Render the 3d models normally\n2 - Render the 3d models using only the edge lines\n3 - Render the 3d models as point clouds\n```\n\n### Other:\n```\nESC - Quit the visualizer\nRight - Go to the next frame\nLeft - Go to the previous frame\nUp - Go to the next 100 frame\nDown - Go to the previous 100 frame\nSpace - Toggle frame auto-changing\nF12 - Toggle exporting the visualized frame to file\n```\n\n", "description_content_type": "text/markdown", "docs_url": null, "download_url": "", "downloads": { "last_day": -1, "last_month": -1, "last_week": -1 }, "home_page": "https://github.com/NVIDIA/Dataset_Utilities", "keywords": "nvdu,nvidia", "license": "Creative Commons Attribution-NonCommercial-ShareAlike 4.0. https://creativecommons.org/licenses/by-nc-sa/4.0/", "maintainer": "Thang To", "maintainer_email": "thangt@nvidia.com", "name": "nvdu", "package_url": "https://pypi.org/project/nvdu/", "platform": "", "project_url": "https://pypi.org/project/nvdu/", "project_urls": { "Homepage": "https://github.com/NVIDIA/Dataset_Utilities" }, "release_url": "https://pypi.org/project/nvdu/1.0.0.0/", "requires_dist": [ "numpy", "opencv-python", "pyrr", "PyWavefront (==0.2.0)", "pyglet", "fuzzyfinder" ], "requires_python": "", "summary": "Nvidia Dataset Utilities", "version": "1.0.0.0" }, "last_serial": 4158029, "releases": { "1.0.0.0": [ { "comment_text": "", "digests": { "md5": "bf825695379bd3aae27b9035e1830981", "sha256": "afbf0ab34a5a55f5fbe20d0c0edc42a392570c3f397e02f8cae2951b9c9a47bd" }, "downloads": -1, "filename": "nvdu-1.0.0.0-py2.py3-none-any.whl", "has_sig": false, "md5_digest": "bf825695379bd3aae27b9035e1830981", "packagetype": "bdist_wheel", "python_version": "py2.py3", "requires_python": null, "size": 46533, "upload_time": "2018-08-10T18:45:44", "url": "https://files.pythonhosted.org/packages/7d/c0/d5b08d4268afbe1507b94b3ea9f913c16bb8821c097e834941a0d38648be/nvdu-1.0.0.0-py2.py3-none-any.whl" }, { "comment_text": "", "digests": { "md5": "9b09731299073d49bb22be9e28dd63ad", "sha256": "8fec5784a5acf3eb45c4f4b297cafb54a43855534d65edf0c24467faae80f7be" }, "downloads": -1, "filename": "nvdu-1.0.0.0.tar.gz", "has_sig": false, "md5_digest": "9b09731299073d49bb22be9e28dd63ad", "packagetype": "sdist", "python_version": "source", "requires_python": null, "size": 28359, "upload_time": "2018-08-10T18:45:46", "url": "https://files.pythonhosted.org/packages/83/a0/6d0c1017faf750a3c14bce2b3d5fe42950f397f5c26102c2717a21606d8b/nvdu-1.0.0.0.tar.gz" } ] }, "urls": [ { "comment_text": "", "digests": { "md5": "bf825695379bd3aae27b9035e1830981", "sha256": "afbf0ab34a5a55f5fbe20d0c0edc42a392570c3f397e02f8cae2951b9c9a47bd" }, "downloads": -1, "filename": "nvdu-1.0.0.0-py2.py3-none-any.whl", "has_sig": false, "md5_digest": "bf825695379bd3aae27b9035e1830981", "packagetype": "bdist_wheel", "python_version": "py2.py3", "requires_python": null, "size": 46533, "upload_time": "2018-08-10T18:45:44", "url": "https://files.pythonhosted.org/packages/7d/c0/d5b08d4268afbe1507b94b3ea9f913c16bb8821c097e834941a0d38648be/nvdu-1.0.0.0-py2.py3-none-any.whl" }, { "comment_text": "", "digests": { "md5": "9b09731299073d49bb22be9e28dd63ad", "sha256": "8fec5784a5acf3eb45c4f4b297cafb54a43855534d65edf0c24467faae80f7be" }, "downloads": -1, "filename": "nvdu-1.0.0.0.tar.gz", "has_sig": false, "md5_digest": "9b09731299073d49bb22be9e28dd63ad", "packagetype": "sdist", "python_version": "source", "requires_python": null, "size": 28359, "upload_time": "2018-08-10T18:45:46", "url": "https://files.pythonhosted.org/packages/83/a0/6d0c1017faf750a3c14bce2b3d5fe42950f397f5c26102c2717a21606d8b/nvdu-1.0.0.0.tar.gz" } ] }