OpenCV Python - 用背景颜色替换蒙版

Question

我想从图像中删除蒙版并将其替换为背景色。在这种情况下，我想从原始图像中删除克里斯蒂亚诺的脸

这是原始图像

这是面具

这是我现在使用的代码

# load background (could be an image too)
bk = np.full(frame.shape, 255, dtype=np.uint8)  # white bk


# blur the mask to help remove noise, then apply the
# mask to the frame
skinMask = cv2.GaussianBlur(skinMask, (3, 3), 0)



#skinMask = cv2.bitwise_not(skinMask)
skin = cv2.bitwise_and(frame, frame, mask = skinMask)

cv2.imshow("mask", skin)

mask = cv2.bitwise_not(skinMask)
bk_masked = cv2.bitwise_and(bk, bk, mask=mask)

# combine masked foreground and masked background 
final = cv2.bitwise_or(bk_masked,skin)

Answer 1

我不确定面具的形状究竟是什么，但交换操作应该非常直接。

# Assuming frame is of shape (531, 403, 3)
# And skinMask is of shape (527, 401, 3)
# Which is what it is when you do an cv2.imread on your posted images

# First find all the coordinates you want to swap
faceCoords = np.where(skinMask != [0,0,0])  # 0,0,0 is black

# Then swap the values from your bk image like so
frame[faceCoords[0],faceCoords[1],faceCoords[2]] = bk[faceCoords[0],faceCoords[1],faceCoords[2]]

这应生成一个图像，背景值替换蒙版值的正部分。