File size: 3,567 Bytes
5767e72 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 | import argparse
import torch
import os, sys
from datetime import datetime
from competition_utils import transform_to_videos
dir_path = os.path.dirname(os.path.realpath(__file__))
parent_dir_path = os.path.abspath(os.path.join(dir_path, os.pardir))
sys.path.append(parent_dir_path)
from competitions import VBenchCompetition
def parse_args():
parser = argparse.ArgumentParser()
parser.add_argument(
"--submission_path",
type=str,
required=False,
help="folder that contains short videos or long videos"
)
parser.add_argument(
"--frame_rate",
type=int,
required=False,
help="frame rate of generated videos"
)
parser.add_argument(
"--video_path",
type=str,
required=False,
help="folder that contains the sampled videos"
)
parser.add_argument(
"--output_path",
type=str,
default="./evaluate_results",
help="output path that save evaluation results"
)
parser.add_argument(
"--dimension",
nargs='+',
required=True,
help="list of evaluation dimensions, usage: --dimension <dim_1> <dim_2>",
)
parser.add_argument(
"--prompt_file",
type=str,
required=True,
default="./short_prompt_list.txt",
help="Specify the path of the file that contains prompt lists"
)
args = parser.parse_args()
return args
def main():
args = parse_args()
print(f"args: {args}")
if not args.video_path:
### transform png frames to mp4 video
assert args.submission_path is not None and args.frame_rate is not None, "You need to provide the submission_path\
and the frame rate for generating the video."
args.video_path = os.path.join(args.output_path, "evaluated_videos")
transform_to_videos(args.submission_path, args.video_path, args.frame_rate)
device = torch.device("cuda")
myvbench = VBenchCompetition(device, None, args.output_path)
print(f'start evaluation')
current_time = datetime.now().strftime('%Y-%m-%d-%H:%M:%S')
kwargs = {
'imaging_quality_preprocessing_mode': 'longer'
}
with open(args.prompt_file, "r") as f:
prompts = [line.strip() for line in f.readlines()]
if "short_prompt_list" in args.prompt_file:
myvbench.evaluate(
videos_path = args.video_path,
name = f'results_short_{current_time}',
prompt_list=prompts,
dimension_list = args.dimension,
**kwargs
)
elif "long_prompt_list" in args.prompt_file:
kwargs['sb_clip2clip_feat_extractor'] = 'dino'
kwargs['bg_clip2clip_feat_extractor'] = 'clip'
kwargs['clip_length_config'] = "clip_length_mix.yaml"
kwargs['w_inclip'] = 1.0
kwargs['w_clip2clip'] = 0.0
kwargs['use_semantic_splitting'] = True
kwargs['slow_fast_eval_config'] = "configs/slow_fast_params.yaml"
kwargs['dev_flag'] = False
kwargs['sb_mapping_file_path'] = "configs/subject_mapping_table.yaml"
kwargs['bg_mapping_file_path'] = "configs/background_mapping_table.yaml"
myvbench.evaluate_long(
videos_path = args.video_path,
name = f'results_long_{current_time}',
prompt_list=prompts,
dimension_list = args.dimension,
**kwargs
)
print("done")
if __name__ == "__main__":
main()
|