Unofficial PyTorch implementation of PointNetVlad (https://github.com/mikacuy/pointnetvlad)
I kept almost everything not related to tensorflow as the original implementation. The main differences are:
- Multi-GPU support
- Configuration file (config.py)
- Evaluation on the eval dataset after every epochs
This implementation achieved an average top 1% recall on oxford baseline of 84.81%
- PyTorch 0.4.0
- tensorboardX
cd generating_queries/
# For training tuples in our baseline network
python generate_training_tuples_baseline.py
# For training tuples in our refined network
python generate_training_tuples_refine.py
# For network evaluation
python generate_test_sets.py
python train_pointnetvlad.py --dataset_folder $DATASET_FOLDER
python evaluate.py --dataset_folder $DATASET_FOLDER
Take a look at train_pointnetvlad.py and evaluate.py for more parameters