(物体を表現した語句に対する3Dデータの生成)
This requires Kaolin(https://github.com/NVIDIAGameWorks/kaolin) pytorch packages.
Environments: https://github.com/herozxzx/Text_to_Voxel/blob/master/setup.txt
This guide helps you set up a Kaolin environment on Ubuntu 20.04 using Docker, NVIDIA tools, and Kaolin library for 3D deep learning.
- Docker installed with NVIDIA support
- NVIDIA driver and toolkit installed
- X server configured for GUI in Docker
Use the provided Docker image for Kaolin environment:
docker run -it -v /tmp/.X11-unix:/tmp/.X11-unix -v /mnt/ShareFiles:/mnt/ShareFiles -e DISPLAY=unix$DISPLAY --net=host --gpus all --cap-add=SYS_PTRACE --security-opt seccomp=unconfined nvidia/cudagl:10.1-devel /bin/bash