import cv2 import matplotlib.pyplot as plt import copy import numpy as np from PIL import Image import os from src import util from src.body import Body # from src.infer_video_d2 import infer_image,load_predictor from video_pose.inference.infer_video_d2 import run_3d from video_pose.data.prepare_data_2d_custom import custom from video_pose.run import run_video_reconstruction def run_3d_for_video(original_video_path): run_3d(original_video_path) video_analyzed_path = original_video_path + ".npz" print(video_analyzed_path) custom(video_analyzed_path) compressed_analyzed_path = video_analyzed_path.replace(".npz", "_compressed.npz") run_video_reconstruction(original_video_path, compressed_analyzed_path) reconstructed_video_path = original_video_path.replace(".mp4", "_reconstructed.mp4") print(reconstructed_video_path) # return reconstructed_video_path def run_openpose_for_image_front(index): PNG_OR_NOT = True # hand_estimation = Hand('model/hand_pose_model.pth') test_image = 'pose_source_images/{}-0.png'.format(index) if os.path.exists(test_image) is False: PNG_OR_NOT = False test_image = 'pose_source_images/{}-0.jpg'.format(index) oriImg = cv2.imread(test_image) # B,G,R order im = cv2.imread(test_image, cv2.IMREAD_UNCHANGED) # --- for openpose --- body_estimation = Body('model/body_pose_model.pth') candidate, subset = body_estimation(oriImg) # --- for detectron --- # predictor = load_predictor() # candidate, subset = infer_image(oriImg, predictor) np.save('pose_temp_data/{}-0.npy'.format(index), candidate) canvas = copy.deepcopy(oriImg) canvas = util.draw_bodypose(canvas, candidate, subset) # sssssssssssss if PNG_OR_NOT == True: alpha_value_new = np.reshape(im[:, :, 3], (im.shape[0], im.shape[1], 1)) canvas = np.c_[canvas, alpha_value_new] # test = np.c_[im[:,:,0:3],alpha_value_new] # print(test) # print(im) # cv2.imwrite('pose_processed_images/tmp_transparent_ori.png', oriImg, [int(cv2.IMWRITE_PNG_COMPRESSION), 9]) cv2.imwrite('pose_processed_images/{}-0-result.png'.format(index), canvas, [int(cv2.IMWRITE_PNG_COMPRESSION), 9]) # cv2.imwrite('pose_processed_images/tmp_transparent_test.png', test, [int(cv2.IMWRITE_PNG_COMPRESSION), 9]) # cv2.imwrite('pose_processed_images/tmp_transparent_im.png', im, [int(cv2.IMWRITE_PNG_COMPRESSION), 9]) # plt.imshow(canvas[:, :, [2, 1, 0]]) # plt.axis('off') # plt.savefig('pose_processed_images/{}-1-result.png'.format(index)) # plt.show() def run_openpose_for_image_side(index): PNG_OR_NOT = True body_estimation = Body('model/body_pose_model.pth') # hand_estimation = Hand('model/hand_pose_model.pth') test_image = 'pose_source_images/{}-1.png'.format(index) if os.path.exists(test_image) is False: PNG_OR_NOT = False test_image = 'pose_source_images/{}-1.jpg'.format(index) oriImg = cv2.imread(test_image) # B,G,R order im = cv2.imread(test_image, cv2.IMREAD_UNCHANGED) # --- for openpose --- body_estimation = Body('model/body_pose_model.pth') candidate, subset = body_estimation(oriImg) # --- for detectron --- # predictor = load_predictor() # candidate, subset = infer_image(oriImg, predictor) np.save('pose_temp_data/{}-1.npy'.format(index), candidate) canvas = copy.deepcopy(oriImg) canvas = util.draw_bodypose(canvas, candidate, subset) if PNG_OR_NOT == True: alpha_value_new = np.reshape(im[:, :, 3], (im.shape[0], im.shape[1], 1)) canvas = np.c_[canvas, alpha_value_new] # test = np.c_[im[:,:,0:3],alpha_value_new] # print(test) # print(im) # cv2.imwrite('pose_processed_images/tmp_transparent_ori.png', oriImg, [int(cv2.IMWRITE_PNG_COMPRESSION), 9]) cv2.imwrite('pose_processed_images/{}-1-result.png'.format(index), canvas, [int(cv2.IMWRITE_PNG_COMPRESSION), 9]) # cv2.imwrite('pose_processed_images/tmp_transparent_test.png', test, [int(cv2.IMWRITE_PNG_COMPRESSION), 9]) # cv2.imwrite('pose_processed_images/tmp_transparent_im.png', im, [int(cv2.IMWRITE_PNG_COMPRESSION), 9]) # plt.imshow(canvas[:, :, [2, 1, 0]]) # plt.axis('off') # plt.savefig('pose_processed_images/{}-1-result.png'.format(index)) # plt.show() def run_openpose_for_normal(index,name): print(index) # 输入index和原来一样，就是序号，你可以更改一下输入或者输出，方便你们处理 # 判断是否是PNG格式，如果是，增添对alpha通道的处理 PNG_OR_NOT = True # 调用的模型路径，无需更改 body_estimation = Body('model/body_pose_model.pth') #原图像的位置，需要更改 ！！ p='./capture_image/'+name test_image = p+'/capture_image{}.png'.format(index) if os.path.exists(test_image) is False: PNG_OR_NOT = False # 原图像的位置，需要更改 ！！ test_image = p+'/capture_image{}.jpg'.format(index) oriImg = cv2.imread(test_image) # B,G,R order im = cv2.imread(test_image, cv2.IMREAD_UNCHANGED) candidate, subset = body_estimation(oriImg) # 坐标点数值保存路径，需要更改！！ np.save(p+'/capture_image{}-1.png'.format(index), candidate) print(p+'/capture_image{}.png'.format(index)) # 处理 canvas = copy.deepcopy(oriImg) canvas = util.draw_bodypose(canvas, candidate, subset) #if PNG_OR_NOT == True: #alpha_value_new = np.reshape(im[:,:,3],(im.shape[0],im.shape[1],1)) #canvas = np.c_[canvas, alpha_value_new] # 结果图片的保存路径，需要更改 ！！ cv2.imwrite(p+'/capture_image_result{}.png'.format(index), canvas, [int(cv2.IMWRITE_PNG_COMPRESSION), 9]) if __name__ == '__main__': # run_openpose_for_normal(1) input_video = "video/15.mp4" run_3d_for_video(input_video)