Neural mesh models for 3D reasoning.
Project description
Neural mesh models for 3D reasoning.
Features
Models
Original NeMo (3D pose estimation)
3D aware classification NeMo (3D pose estimation and classification)
6D pose estimation NeMo
(Deformable NeMo ?)
Baselines
ResNet50
Starmap
Transformer
Faster RCNN
Datasets
OOD-CV
Pascal3D+
(Synthetic data ?)
(ObjectNet ?)
Requirements
Python Environment
Create
conda
environment:
conda create -n nemo python=3.9
conda activate nemo
Install
PyTorch
(see pytorch.org):
conda install pytorch==1.12.0 torchvision==0.13.0 torchaudio==0.12.0 cudatoolkit=10.2 -c pytorch
Install
PyTorch3D
(see github.com/facebookresearch/pytorch3d):
conda install -c fvcore -c iopath -c conda-forge fvcore iopath
conda install -c bottler nvidiacub
conda install pytorch3d -c pytorch3d
Install other dependencies:
conda install numpy matplotlib scipy scikit-image
conda install pillow
conda install -c conda-forge timm tqdm pyyaml transformers
pip install wget BboxTools opencv-python
Install NeMo-Extensions:
pip install -e .
Data Preparation
See data/README.
TODO
[ ] Add support for VoGe Renderer
[ ] Add support for multiple GPUs support
[ ] Add support for some visualizations
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
neural_mesh_model-1.0.0.tar.gz
(40.4 kB
view hashes)