INFO: Created TensorFlow Lite XNNPACK delegate for CPU. Traceback (most recent call last): File "counter.py", line 51, in <module> results = pose.process(image) File "/usr/local/lib/python3.7/dist-packages/mediapipe/python/solutions/pose.py", line 185, in process results = super().process(input_data={'image': image}) File "/usr/local/lib/python3.7/dist-packages/mediapipe/python/solution_base.py", line 338, in process if data.shape[2] != RGB_CHANNELS: AttributeError: 'tuple' object has no attribute 'shape'报错

这个错误的原因可能是您传递给`process()`函数的输入数据格式不正确。请确保您传递给函数的图像数据是一个三维数组，具有RGB通道。您可以尝试将输入图像转换为正确的格式，并确保它是一个numpy数组，例如：

import numpy as np

# 将图像转换为 numpy 数组
image = np.array(image)

# 确保图像具有三个通道
if image.ndim == 2:
    image = np.tile(np.expand_dims(image, axis=2), (1, 1, 3))
elif image.shape[2] == 4:
    image = image[:, :, :3]

然后您可以将转换后的图像传递给`process()`函数进行处理。