# Draw predicting score for only 1 person (not using for)
if DO_INFER_ACTIONS and len(dict_id2skeleton):
multipeople_classifier.get(
id='min').draw_scores_onto_image(image_disp)
# -- Write skeleton.txt and image.png
if SAVE_RESULTANT_SKELETON_TO_TXT_AND_IMAGE:
ids = sorted(dict_id2skeleton.keys())
skel_to_save = [
img_info + dict_id2skeleton[id].tolist() for id in ids
]
myio.save_skeletons(
SAVE_DETECTED_SKELETON_TO + myfunc.int2str(ith_img, 5) +
".txt", skel_to_save)
cv2.imwrite(
SAVE_DETECTED_SKELETON_IMAGES_TO + myfunc.int2str(ith_img, 5) +
".png", image_disp)
if FROM_TXTSCRIPT or FROM_WEBCAM: # Save source image
cv2.imwrite(
SAVE_DETECTED_SKELETON_IMAGES_TO +
myfunc.int2str(ith_img, 5) + "_src.png", img)
# -- Display
if 1:
if ith_img == 1:
window_name = "action_recognition"
cv2.namedWindow(window_name, cv2.WND_PROP_FULLSCREEN)
print("prediced label is :", prediced_label)
else:
prediced_label = action_type
print("Ground_truth label is :", prediced_label)
if 1:
# Draw skeleton
if ith_skel == 0:
my_detector.draw(image_disp, humans)
# Draw bounding box and action type
drawActionResult(image_disp, skeleton, prediced_label)
# Write result to txt/png
if SAVE_RESULTANT_SKELETON_TO_TXT_AND_IMAGE:
myio.save_skeletons(SAVE_DETECTED_SKELETON_TO
+ int2str(ith_img, 5)+".txt", skelsInfo)
cv2.imwrite(SAVE_DETECTED_SKELETON_IMAGES_TO
+ int2str(ith_img, 5)+".png", image_disp)
if 1: # Display
cv2.imshow("action_recognition",
cv2.resize(image_disp,(0,0),fx=1.5,fy=1.5))
q = cv2.waitKey(1)
if q!=-1 and chr(q) == 'q':
break
# Loop
print("\n")
ith_img += 1