human-3d / Human3D

We propose the first multi-human body-part segmentation model, called Human3D 🧑‍🤝‍🧑, that directly operates on 3D scenes. In an extensive analysis, we validate the benefits of training on synthetic data on multiple baselines and tasks.
https://human-3d.github.io/
54 stars 6 forks source link

How to predict my data #7

Open ZhYuLn opened 12 months ago

ZhYuLn commented 12 months ago

Hi: This is my directory tree in the project Human3D:

├── benchmark
│   └── __pycache__
├── checkpoints
├── conf
│   ├── augmentation
│   ├── callbacks
│   ├── data
│   ├── full
│   ├── logging
│   ├── loss
│   ├── matcher
│   ├── metrics
│   ├── model
│   ├── optimizer
│   ├── scheduler
│   └── trainer
├── data
│   ├── processed
│   └── raw
├── datasets
│   └── preprocessing
├── docs
│   └── assets
├── models
│   ├── __pycache__
│   ├── metrics
│   └── modules
├── occlusion_subsets
├── outputs
│   └── 2023-12-03
├── saved
│   └── hydra_logs
├── scripts
│   ├── eval
│   └── train
├── third_party
│   ├── MinkowskiEngine
│   └── pointnet2
├── trainer
│   └── __pycache__
└── utils
    ├── __pycache__
    ├── pointops2
    └── votenet_utils

which directory I should put my ply file, and if the eval_human3d.sh worked, which diretory will saved the predicted ply.

jk456445 commented 11 months ago

Hello, did you find the corrrect way to predict your own data? I have the save question, too.

AT181903 commented 9 months ago

Is there a script to predict using my own data?