This is a 6D pose network based on PVNet
- configs - Some setting are configured here
- datasets - LINEMOD dataset
- evaluator - The evaluation process is implemented here
- log_info - Used for storing models and results
- nets - The network model is implemented here
- trainer - The supplementary class for training the network
- utils
- decorators - Helper decorators
- io_utils - IO Helper
- refinement - ICP refinement process is here
- visual_utils - Visualization helper
There are some steps you need to finish before running the program.
- Specify the path of the dataset in your own machine in configs/constants.py
- Specify teh path for saving the results/model in configs/constants.py
- Specify the training configurations in configs/training_configs.py