Open Ha-kki opened 3 years ago
well, for running the training, just follow the instruction in the prepare the training data section and use the command suggested, then it will automatically extract the data structures from the datasets for training:
python data/prepare_train_data.py --dataset_dir=/path/to/raw/kitti/dataset/ --dataset_name='kitti_raw_eigen' --dump_root=/path/to/resulting/formatted/data/ --seq_length=3 --img_width=416 --img_height=128 --num_threads=4
I have downloaded the kitti raw datasets from its official website, but I don't know the structure of its files which is required in this code. Could anyone share structures of these datasets used in this program. Thanks!:)