Download temporal_segmentation.py from X-NS/NSAQA: direct link, hf CLI and curl.
- Browser
- Download file 11.7 kB
-
https://huggingface.co/spaces/X-NS/NSAQA/resolve/main/temporal_segmentation.py
- Command line
-
hf download hf://spaces/X-NS/NSAQA/temporal_segmentation.py
-
curl -L -o temporal_segmentation.py https://huggingface.co/spaces/X-NS/NSAQA/resolve/main/temporal_segmentation.py
11.7 kB
| from microprograms.temporal_segmentation.entry import entry_microprogram_one_frame | |
| from microprograms.temporal_segmentation.somersault import somersault_microprogram_one_frame | |
| from microprograms.temporal_segmentation.twist import twist_microprogram_one_frame | |
| from microprograms.temporal_segmentation.start_takeoff import takeoff_microprogram_one_frame | |
| from microprograms.errors.distance_from_springboard_micro_program import board_end | |
| from microprograms.errors.splash_micro_program import get_splash_from_one_frame | |
| from microprograms.errors.distance_from_springboard_micro_program import calculate_distance_from_springboard_for_one_frame | |
| from microprograms.errors.distance_from_springboard_micro_program import calculate_distance_from_platform_for_one_frame | |
| from microprograms.errors.angles_micro_programs import applyFeetApartError | |
| from microprograms.errors.angles_micro_programs import applyPositionTightnessError | |
| from models.detectron2.platform_detector_setup import get_platform_detector | |
| from models.pose_estimator.pose_estimator_model_setup import get_pose_estimation | |
| from models.detectron2.diver_detector_setup import get_diver_detector | |
| from models.pose_estimator.pose_estimator_model_setup import get_pose_model | |
| from models.detectron2.splash_detector_setup import get_splash_detector | |
| from somersault_counter import som_counter, twist_counter | |
| from microprograms.errors.over_rotation import over_rotation | |
| import pickle | |
| import os | |
| import math | |
| import numpy as np | |
| import cv2 | |
| # returns None if either pose_pred or board_end_coord is None | |
| # returns True if diver is on board, returns False if diver is off board | |
| def detect_on_board(board_end_coord, board_side, pose_pred, handstand): | |
| if pose_pred is None: | |
| print("pose_pred is None") | |
| return | |
| if board_end_coord is None: | |
| print("board end coord is None") | |
| return | |
| # side = find_which_side_board_on(outputs) | |
| if board_side == 'left': | |
| # if right of board end | |
| if np.array(pose_pred)[0][2][0] > int(board_end_coord[0]): | |
| # print("board on left side, we're saying that the diver is on the right") | |
| # if standing_dive: | |
| if handstand: | |
| # print("board end to wrist dist:", math.dist(np.array(pose_pred)[0][15], board_end_coord)) | |
| # print("wrist to elbow dist:", math.dist(np.array(pose_pred)[0][14], np.array(pose_pred)[0][15])) | |
| distance = math.dist(np.array(pose_pred)[0][15], board_end_coord) < math.dist(np.array(pose_pred)[0][14], np.array(pose_pred)[0][15]) * 1.5 | |
| else: | |
| # print("board end to ankle dist:", math.dist(np.array(pose_pred)[0][5], board_end_coord)) | |
| # print("ankle to knee dist:", math.dist(np.array(pose_pred)[0][5], np.array(pose_pred)[0][4])) | |
| distance = math.dist(np.array(pose_pred)[0][5], board_end_coord) < math.dist(np.array(pose_pred)[0][5], np.array(pose_pred)[0][4]) * 1.5 | |
| # print("distance", distance) | |
| return distance | |
| # return False | |
| # if left of board end | |
| else: | |
| # print("board on left side, we're saying that the diver is on the left") | |
| return True | |
| else: | |
| # if right of board end | |
| if np.array(pose_pred)[0][2][0] > int(board_end_coord[0]): | |
| # print("board on right side, we're saying that the diver is on the right") | |
| return True | |
| # if left of board end | |
| else: | |
| # print("board on right side, we're saying that the diver is on the left") | |
| # if standing_dive: | |
| if handstand: | |
| distance = math.dist(np.array(pose_pred)[0][10], board_end_coord) < math.dist(np.array(pose_pred)[0][11], np.array(pose_pred)[0][10]) * 1.5 | |
| else: | |
| distance = math.dist(np.array(pose_pred)[0][0], board_end_coord) < math.dist(np.array(pose_pred)[0][1], np.array(pose_pred)[0][0]) * 1.5 | |
| # print("distance", distance) | |
| return distance | |
| # return False | |
| def main(): | |
| first_folder = input("what is the first folder? Ex: 01, FINAWorldChampionships2019_Women10m_final_r1, etc. ") | |
| second_folder = input("what is the second folder? (dive within the first folder)") | |
| takeoff = [] | |
| twist = [] | |
| som = [] | |
| entry = [] | |
| distance_from_board = [] | |
| position_tightness = [] | |
| feet_apart = [] | |
| splash = [] | |
| above_board = True | |
| board_side = input("what side is the board on? type either 'left' or 'right'") | |
| on_board = True | |
| handstand = input("is the dive a handstand dive? type either 'True' or 'False' ") == 'True' | |
| expected_twists = int(input("what are the expected number of twists? type 1 for half a twist, 2 for full twist, etc.")) | |
| expected_som = int(input("what are the expected number of somersaults? type 1 for half a somersault, 2 for full somersault, etc.")) | |
| # expected_twists= 0 | |
| # expected_som = 7 | |
| platform_detector = get_platform_detector() | |
| splash_detector = get_splash_detector() | |
| diver_detector = get_diver_detector() | |
| pose_model = get_pose_model() | |
| dive_folder_num = "{}_{}".format(first_folder, second_folder) | |
| directory = './FineDiving/datasets/FINADiving_MTL_256s/{}/{}/'.format(first_folder, second_folder) | |
| # dive_folder_num = '01_1' | |
| # directory = './FineDiving/datasets/FINADiving_MTL_256s/01/1/' | |
| # dive_folder_num = 'FINAWorldChampionships2019_Women10m_final_r1_0' | |
| # directory = './FineDiving/datasets/FINADiving_MTL_256s/FINAWorldChampionships2019_Women10m_final_r1/0/' | |
| file_names = os.listdir(directory) | |
| # with open('./output/joint_plots/{}/pose_preds.pkl'.format(dive_folder_num), 'rb') as pickle_file: | |
| # pose_preds = pickle.load(pickle_file) | |
| j = 0 | |
| # if len(pose_preds) > len(file_names): | |
| # print("WRONG POSE_PREDS") | |
| prev_pred = None | |
| som_prev_pred =None | |
| half_som_count=0 | |
| petal_count = 0 | |
| in_petal = False | |
| for i in range(len(file_names)): | |
| # pose_pred = None | |
| filepath = directory + file_names[i] | |
| print("filepath:", filepath) | |
| if file_names[i][-4:] != ".jpg": | |
| continue | |
| diver_box, pose_pred = get_pose_estimation(filepath, diver_detector=diver_detector, pose_model=pose_model) | |
| # if j < len(pose_preds): | |
| # print("filepath has pose_pred:", filepath) | |
| # pose_pred = pose_preds[j] | |
| # j += 1 | |
| calculated_half_som_count, skip = som_counter(pose_pred, som_prev_pred, half_som_count=half_som_count, handstand=handstand) | |
| if not skip: | |
| som_prev_pred = pose_pred | |
| print("calculated_half_som_count:", calculated_half_som_count) | |
| calculated_petal_count, calculated_in_petal = twist_counter(pose_pred, prev_pose_pred=prev_pred, in_petal=in_petal, petal_count=petal_count) | |
| print("calculated_petal_count", calculated_petal_count) | |
| print("calculated_in_petal", calculated_in_petal) | |
| im = cv2.imread(filepath) | |
| outputs = platform_detector(im) | |
| board_end_coord = board_end(outputs, board_side=board_side) | |
| # if board_end_coord is None: | |
| # print("NO BOARD NONE CRYING") | |
| if above_board and not on_board and board_end_coord is not None and pose_pred is not None and np.array(pose_pred)[0][2][1] > int(board_end_coord[1]): | |
| above_board=False | |
| if on_board and detect_on_board(board_end_coord, board_side, pose_pred, handstand) is not None and not detect_on_board(board_end_coord, board_side, pose_pred, handstand): | |
| on_board = False | |
| print('ON_BOARD:', on_board) | |
| print('ABOVE_BOARD:', above_board) | |
| calculated_takeoff = takeoff_microprogram_one_frame(filepath, above_board=above_board, on_board=on_board, pose_pred=pose_pred) | |
| calculated_twist = twist_microprogram_one_frame(filepath, on_board=on_board, pose_pred=pose_pred, expected_twists=expected_twists, petal_count=petal_count, expected_som=expected_som, half_som_count=half_som_count) | |
| calculated_som = somersault_microprogram_one_frame(filepath, pose_pred=pose_pred, on_board=on_board, expected_som=expected_som, half_som_count=half_som_count, expected_twists=expected_twists, petal_count=petal_count) | |
| calculated_entry = entry_microprogram_one_frame(filepath, above_board=above_board, on_board=on_board, pose_pred=pose_pred, expected_twists=expected_twists, petal_count=petal_count, expected_som=expected_som, half_som_count=half_som_count, splash_detector=splash_detector, visualize=True, dive_folder_num=dive_folder_num) | |
| if calculated_takeoff == 1: | |
| # distance from board | |
| dist = calculate_distance_from_platform_for_one_frame(filepath, visualize=False, pose_pred=pose_pred, board_end_coord=board_end_coord, platform_detector=platform_detector) # saves photo to ./output/data/distance_from_board/ | |
| distance_from_board.append(dist) | |
| # height off board | |
| # height_off_board.append(dist) | |
| # proximity to the edge of board | |
| # proximity_to_end_board.append(dist) | |
| elif calculated_som == 1: | |
| half_som_count = calculated_half_som_count | |
| # petal_count = calculated_petal_count | |
| # in_petal = calculated_in_petal | |
| if above_board: | |
| dist = calculate_distance_from_platform_for_one_frame(filepath, visualize=False, pose_pred=pose_pred, board_end_coord=board_end_coord, platform_detector=platform_detector) # saves photo to ./output/data/distance_from_board/ | |
| # distance from board | |
| distance_from_board.append(dist) | |
| # rotation speed | |
| # position tightness | |
| position_tightness.append(applyPositionTightnessError(filepath, pose_pred=pose_pred)) | |
| # feet flat | |
| # feet/legs apart | |
| feet_apart.append(applyFeetApartError(filepath, pose_pred=pose_pred)) | |
| elif calculated_twist == 1: | |
| half_som_count = calculated_half_som_count | |
| petal_count = calculated_petal_count | |
| in_petal = calculated_in_petal | |
| if above_board: | |
| dist = calculate_distance_from_platform_for_one_frame(filepath, visualize=False, pose_pred=pose_pred, board_end_coord=board_end_coord, platform_detector=platform_detector) # saves photo to ./output/data/distance_from_board/ | |
| # distance from board | |
| distance_from_board.append(dist) | |
| # twisting speed | |
| # position tightness | |
| position_tightness.append(180 - applyPositionTightnessError(filepath, pose_pred=pose_pred)) | |
| # feet flat | |
| # feet/legs apart | |
| feet_apart.append(applyFeetApartError(filepath, pose_pred=pose_pred)) | |
| elif calculated_entry == 1: | |
| # over/under twisting | |
| # over/under rotating | |
| print("OVER-ROTATION ERROR:", over_rotation(filepath, pose_pred=pose_pred, diver_detector=diver_detector, pose_model=pose_model)) | |
| # not straight during entry | |
| # splash | |
| splash.append(get_splash_from_one_frame(filepath, predictor=splash_detector, visualize=True)) | |
| else: | |
| print('no phase of dive calculated!') | |
| takeoff.append(calculated_takeoff) | |
| twist.append(calculated_twist) | |
| som.append(calculated_som) | |
| entry.append(calculated_entry) | |
| prev_pred = pose_pred | |
| print("takeoff", takeoff) | |
| print("twist", twist) | |
| print("som", som) | |
| print("entry", entry) | |
| print("distance_from_board", distance_from_board) | |
| print("position_tightness", position_tightness) | |
| print("feet_apart", feet_apart) | |
| print("splash", splash) | |
| if __name__ == "__main__": | |
| main() | |