如何解决函数‘imshow’中的“cv2.错误: OpenCV(4.5.4) :-1:错误：(-5:糟糕的参数)”

Question

我正在学习如何使用opencv，但我遇到了这个问题。

from cvzone.HandTrackingModule import HandDetector
import cv2


cap = cv2.VideoCapture("https://192.168.178.49:8080/video")
detector = HandDetector(maxHands=1, detectionCon=0.7)

while True:
    success, img= cap.read()
    
    img = detector.findHands(img) 
    
    cv2.imshow("AI", img)
    cv2.waitKey(1)

导致此错误：

INFO: Created TensorFlow Lite XNNPACK delegate for CPU.
Traceback (most recent call last):
  File "d:\Programming\Arm Code\testhandai.py", line 13, in <module>
    cv2.imshow("AI", img)
cv2.error: OpenCV(4.5.4) :-1: error: (-5:Bad argument) in function 'imshow'
> Overload resolution failed:
>  - mat is not a numerical tuple
>  - Expected Ptr<cv::cuda::GpuMat> for argument 'mat'
>  - Expected Ptr<cv::UMat> for argument 'mat'

我正在使用Python3.8 64位和所有包的最新版本。谢谢。

Answer 1

detector.findHands(img)的输出是一个元组。您应该将它的第二个元素作为cv2.imshow()的输入

from cvzone.HandTrackingModule import HandDetector
import cv2


cap = cv2.VideoCapture("https://192.168.178.49:8080/video")
detector = HandDetector(maxHands=1, detectionCon=0.7)

while True:
    success, img= cap.read()
    
    img = detector.findHands(img) 
    
    cv2.imshow("AI", img[1])
    cv2.waitKey(1)

Answer 2

mediapipe增加了一个新的变量，它破坏了手跟踪和姿态估计、呼叫复杂度和model_complexity。见下文。

现在我正在做人脸检测，模块工作得很好，但是当我把它拉到另一个不同的地方时，它就失败了。

用于手动跟踪模块

def __init__(self, mode=False, maxHands=2, complexity = 1, detectionCon=0.5, trackCon=0.5):
    self.mode = mode
    self.maxHands = maxHands
    self.complexity = complexity
    self.detectionCon = detectionCon
    self.trackCon = trackCon
    self.mpHands = mp.solutions.hands
    self.hands = self.mpHands.Hands(self.mode, self.maxHands, self.complexity,
                                    self.detectionCon, self.trackCon, )

关于员额esitimation模块：

def __init__(self, mode = False, model_complexity = 1, smooth = True,
                    enable_segmentation = False, smooth_segmentation = True, min_detection_confidence = 0.5,
                    min_tracking_confidence = 0.5):
    self.mode = mode
    self.model_complexity = model_complexity
    self.smooth = smooth
    self.enable_segmentation = enable_segmentation
    self.smooth_segmentation = smooth_segmentation
    self.detectionCon = min_detection_confidence
    self.trackCon = min_tracking_confidence
    self.mpDraw = mp.solutions.drawing_utils
    self.mpPose = mp.solutions.pose
    self.pose = self.mpPose.Pose(self.mode, self.model_complexity, self.smooth, self.enable_segmentation,
                                 self.detectionCon, self.trackCon)