{ "id": 117185, "title": "Nvidia's Kaolin for 3D computer vision", "url": "/competitions/3d-object-detection-for-autonomous-vehicles/discussion/117185", "author_name": "", "post_date": "2019-11-13T19:51:33.782702700Z", "votes": 11, "comment_count": 5, "views": 0, "content": "

Nvidia has released a new library specifically focused on 3D computer vision, named as Kaolin.

\n\n
\n

At its core, Kaolin consists of an efficient suite of geometric functions that allow manipulation of 3D content. It can wrap into PyTorch tensors 3D datasets implemented as polygon meshes, point clouds, signed distance functions or voxel grids.

\n \n

With their 3D dataset ready for deep learning, researchers can choose a neural network model from a curated collection that Kaolin supplies. The interface provides a rich repository of models, both baseline and state of the art, for classification, segmentation, 3D reconstruction, super-resolution and more.

\n
\n\n

Here's the official blog: https://news.developer.nvidia.com/kaolin-library-research-3d/\nGithub repository: https://github.com/NVIDIAGameWorks/kaolin/

", "messages": [ { "id": "672353", "postDate": "11/13/2019 19:51:33", "content": "

Nvidia has released a new library specifically focused on 3D computer vision, named as Kaolin.

\n\n
\n

At its core, Kaolin consists of an efficient suite of geometric functions that allow manipulation of 3D content. It can wrap into PyTorch tensors 3D datasets implemented as polygon meshes, point clouds, signed distance functions or voxel grids.

\n \n

With their 3D dataset ready for deep learning, researchers can choose a neural network model from a curated collection that Kaolin supplies. The interface provides a rich repository of models, both baseline and state of the art, for classification, segmentation, 3D reconstruction, super-resolution and more.

\n
\n\n

Here's the official blog: https://news.developer.nvidia.com/kaolin-library-research-3d/\nGithub repository: https://github.com/NVIDIAGameWorks/kaolin/

", "rawMarkdown": "Nvidia has released a new library specifically focused on 3D computer vision, named as `Kaolin`.\n\n> At its core, Kaolin consists of an efficient suite of geometric functions that allow manipulation of 3D content. It can wrap into PyTorch tensors 3D datasets implemented as polygon meshes, point clouds, signed distance functions or voxel grids.\n\n> With their 3D dataset ready for deep learning, researchers can choose a neural network model from a curated collection that Kaolin supplies. The interface provides a rich repository of models, both baseline and state of the art, for classification, segmentation, 3D reconstruction, super-resolution and more.\n\nHere's the official blog: https://news.developer.nvidia.com/kaolin-library-research-3d/\nGithub repository: https://github.com/NVIDIAGameWorks/kaolin/", "votes": null }, { "id": "672473", "postDate": "11/13/2019 23:42:54", "content": "

Great! Thanks for sharing!

", "rawMarkdown": "Great! Thanks for sharing!", "votes": null }, { "id": "672937", "postDate": "11/14/2019 10:03:00", "content": "

Right, so now we have to covert data from nuScenes to KITTI and then to Kaolin, and then convert the predictions all the way back :)

", "rawMarkdown": "Right, so now we have to covert data from nuScenes to KITTI and then to Kaolin, and then convert the predictions all the way back :)", "votes": null }, { "id": "672991", "postDate": "11/14/2019 11:15:49", "content": "

looks similar to tensorflow-graphics https://github.com/tensorflow/graphics

", "rawMarkdown": "looks similar to tensorflow-graphics https://github.com/tensorflow/graphics", "votes": null }, { "id": "673160", "postDate": "11/14/2019 15:43:49", "content": "

great

", "rawMarkdown": "great", "votes": null }, { "id": "1958063", "postDate": "09/27/2022 09:09:01", "content": "

Interesting !

", "rawMarkdown": "Interesting !", "votes": null } ], "comments": [ { "id": 1958063, "author_name": "tea61ketaninamdar", "author_url": "", "post_date": "09/27/2022 09:09:01", "content": "

Interesting !

", "votes": null, "replies": [] }, { "id": 672473, "author_name": "phunghieu", "author_url": "", "post_date": "11/13/2019 23:42:54", "content": "

Great! Thanks for sharing!

", "votes": null, "replies": [] }, { "id": 672937, "author_name": "artyomp", "author_url": "", "post_date": "11/14/2019 10:03:00", "content": "

Right, so now we have to covert data from nuScenes to KITTI and then to Kaolin, and then convert the predictions all the way back :)

", "votes": null, "replies": [] }, { "id": 672991, "author_name": "oarphme", "author_url": "", "post_date": "11/14/2019 11:15:49", "content": "

looks similar to tensorflow-graphics https://github.com/tensorflow/graphics

", "votes": null, "replies": [] }, { "id": 673160, "author_name": "harishjonnada", "author_url": "", "post_date": "11/14/2019 15:43:49", "content": "

great

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "672353": "Nvidia has released a new library specifically focused on 3D computer vision, named as `Kaolin`.\n\n> At its core, Kaolin consists of an efficient suite of geometric functions that allow manipulation of 3D content. It can wrap into PyTorch tensors 3D datasets implemented as polygon meshes, point clouds, signed distance functions or voxel grids.\n\n> With their 3D dataset ready for deep learning, researchers can choose a neural network model from a curated collection that Kaolin supplies. The interface provides a rich repository of models, both baseline and state of the art, for classification, segmentation, 3D reconstruction, super-resolution and more.\n\nHere's the official blog: https://news.developer.nvidia.com/kaolin-library-research-3d/\nGithub repository: https://github.com/NVIDIAGameWorks/kaolin/", "672473": "Great! Thanks for sharing!", "672937": "Right, so now we have to covert data from nuScenes to KITTI and then to Kaolin, and then convert the predictions all the way back :)", "672991": "looks similar to tensorflow-graphics https://github.com/tensorflow/graphics", "673160": "great", "1958063": "Interesting !" }, "source": "meta" }