Question about CO3DV2 Preprocessing #100
Description
def get_parser():
parser = argparse.ArgumentParser()
parser.add_argument("--category", type=str, default=None)
parser.add_argument('--single_sequence_subset', default=False, action='store_true',
help="prepare the single_sequence_subset instead.")
parser.add_argument("--output_dir", type=str, default="data/co3d_processed")
parser.add_argument("--co3d_dir", type=str, required=True)
parser.add_argument("--num_sequences_per_object", type=int, default=50)
parser.add_argument("--seed", type=int, default=42)
parser.add_argument("--min_quality", type=float, default=0.5, help="Minimum viewpoint quality score.")
parser.add_argument("--img_size", type=int, default=512,
help=("lower dimension will be >= img_size * 3/4, and max dimension will be >= img_size"))
When I use a cut3r script to process the dataset, how should I set these parameters? How is valid_seq.json obtained (I found this file, I'm just curious about its source)?