68 lines
2.5 KiB
Python
68 lines
2.5 KiB
Python
|
import sys
|
||
|
|
import argparse
|
||
|
|
sys.path.append('./src/')
|
||
|
|
from main import *
|
||
|
|
|
||
|
|
def test_skeleton_filter(person_name, pose_mode, test_num, cbr_num, model):
|
||
|
|
filename = merge_skeleton_data('data/skeleton_data/' + person_name + '/' + pose_mode)
|
||
|
|
original_data, estimate_data, estimate_state = simulation_ukf(filename, test_num, cbr_num, model)
|
||
|
|
save_skeleton_data_to_csv(person_name, pose_mode, original_data, estimate_data, estimate_state, model)
|
||
|
|
|
||
|
|
def test_skeleton_draw(person_name, pose_mode, model):
|
||
|
|
original_data, estimate_data = read_skeleton_data_from_csv(person_name, pose_mode, model)
|
||
|
|
skeleton_draw(person_name, pose_mode, model, original_data, estimate_data)
|
||
|
|
|
||
|
|
def test_skeleton(person_name, pose_mode, test_num, cbr_num, onFilter, model, onPlot):
|
||
|
|
if onFilter == 'on':
|
||
|
|
test_skeleton_filter(person_name, pose_mode, test_num, cbr_num, model)
|
||
|
|
if onPlot == 'on':
|
||
|
|
test_skeleton_draw(person_name, pose_mode, model)
|
||
|
|
|
||
|
|
def test_one_person_all_mode(person_name, test_num, cbr_num, onFilter, model, onPlot):
|
||
|
|
folder_name = 'data/skeleton_data/' + person_name
|
||
|
|
dir_list = get_dir_name(folder_name)
|
||
|
|
for pose_mode in dir_list:
|
||
|
|
print(person_name, pose_mode)
|
||
|
|
test_skeleton(person_name, pose_mode, test_num, cbr_num, onFilter, model, onPlot)
|
||
|
|
|
||
|
|
if __name__ == '__main__':
|
||
|
|
parser = argparse.ArgumentParser()
|
||
|
|
parser.add_argument('person_name', type=str,
|
||
|
|
metavar='First_string',
|
||
|
|
help='What is your person name?')
|
||
|
|
parser.add_argument('pose_mode', type=str,
|
||
|
|
metavar='Second_string',
|
||
|
|
help='What is your pose? "*" is all pose')
|
||
|
|
|
||
|
|
parser.add_argument('--filter', type=str, default='on',
|
||
|
|
choices=['on','off'],
|
||
|
|
help='Do you want to plot?')
|
||
|
|
parser.add_argument('--model', type=str, default='ukf',
|
||
|
|
choices=['ukf','kf'],
|
||
|
|
help='which type of filter?')
|
||
|
|
parser.add_argument('--plot', type=str, default='on',
|
||
|
|
choices=['on','off'],
|
||
|
|
help='Do you want to plot?')
|
||
|
|
parser.add_argument('--num', type=int, default=1e9,
|
||
|
|
help='How many tests do you want?')
|
||
|
|
parser.add_argument('--cbr_num', type=int, default=1e9,
|
||
|
|
help='How many calibration tests do you want?')
|
||
|
|
|
||
|
|
args = parser.parse_args()
|
||
|
|
|
||
|
|
person_name = args.person_name
|
||
|
|
pose = args.pose_mode
|
||
|
|
onFilter = args.filter
|
||
|
|
model = args.model
|
||
|
|
onPlot = args.plot
|
||
|
|
test_num = args.num
|
||
|
|
cbr_num = args.cbr_num
|
||
|
|
|
||
|
|
if pose == '*':
|
||
|
|
test_one_person_all_mode(person_name, test_num, cbr_num, onFilter, model, onPlot)
|
||
|
|
else:
|
||
|
|
test_skeleton(person_name, pose, test_num, cbr_num, onFilter, model, onPlot)
|
||
|
|
|
||
|
|
# test_skeleton_filter("jiwon", "crossing_arms_30sec", test_num=10)
|
||
|
|
# test_skeleton_draw("jiwon", "crossing_arms_30sec", plot_3D=True)
|