Closed divastar closed 3 years ago
I tried np.int64( casting : mX = np.int64(np.mean(X)) mY = np.int64(np.mean(Y)) length = np.int64(np.int64(((Y[0] - Y[1]) 2 + (X[0] - X[1]) 2) ** 0.5)) angle = np.int64(math.degrees(math.atan2(Y[0] - Y[1], X[0] - X[1]))/2) stickwidth = np.int64((kp_scores[start_p] + kp_scores[end_p]) + 1)
polygon = cv2.ellipse2Poly((mX,mY), (length, stickwidth), int(angle), 0, 360, 1)
doesn't help
Hi, did you mange to solve it?
No. But it is still works without..
On Thu, 11 Feb 2021 at 6:58 Fred Fang notifications@github.com wrote:
Hi, did you mange to solve it?
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/MVIG-SJTU/AlphaPose/issues/800#issuecomment-777202874, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMB4M4BMWFKOGZGNRD45WTTS6NPV7ANCNFSM4XBWXQBA .
It is weird... You can use the --vis_fast flag to use simple drawing, so you wont meet this problem
Hi . I am using a newest pytorch version . And I am having this typing problem did not managed to solve yet. Someone can help? File "C:\pose estimation\AlphaPose\fn.py", line 214, in vis_frame polygon = cv2.ellipse2Poly((int(mX),int(mY)), (int(length/2), stickwidth), int(angle), 0, 360, 1)TypeError: only integer tensors of a single element can be converted to an index
thank you