如何避免estimateRigidTransform返回无

Question

我一直在尝试使用estimateRigidTransform来确定视频中帧对之间的转换矩阵。我的问题是我输入代码的对几乎有一半返回无值。

我尝试将estimateRigidTransform与整个帧一起使用，现在我要输入点了，结果稍好一些，但仍然不够好

############whole frames

frameBefore = cv2.imread("1.png")
frame = cv2.imread("result_00001.png") 
transform = cv2.estimateRigidTransform(frameBefore,frameafter, False)


###########data points

frameBefore = cv2.imread("1.png")
prev_gray = cv2.cvtColor(frameBefore, cv2.COLOR_BGR2GRAY)
prev_pts = cv2.goodFeaturesToTrack(prev_gray,
                                     maxCorners=200,
                                     qualityLevel=0.01,
                                     minDistance=30,
                                     blockSize=3)
frame = cv2.imread("result_00001.png")
curr_gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY) 

curr_pts, status, err = cv2.calcOpticalFlowPyrLK(prev_gray, curr_gray, prev_pts, None) 
assert prev_pts.shape == curr_pts.shape

idx = np.where(status==1)[0]
prev_pts = prev_pts[idx]
curr_pts = curr_pts[idx]

transform = cv2.estimateRigidTransform(prev_pts, curr_pts, fullAffine=False)

对于43000帧，我得到18000无值。我在做什么错了？