I have this code where I label fingers using MediaPipe and OpenCV:

when i run this code it gives an error of image directory dataset is empty while the directory is valid and its also show content when using another method i think the problem is with gesture recognizer
import os

I’m having problems extracting the pinky landmarks and it gives me an error when I declare the point of the phalanx.

When running the demo of selfie segmentation(https://ai.google.dev/edge/mediapipe/solutions/vision/image_segmenter),
it seems that simple selfie segmentation fails to tell man’s head from chair. Multiclass selfie segmentation works greatly so that I want to implement the model. However, it seems that the mediapipe library does not support any functions for it.

I am trying to get multiple people to draw landmarks using MediaPipe’s Pose with Python3.10.14, but it does not work. The program is forced to exit without error. In other words, it is not trapped by “except” in the code. The following is a code which always fails at detection_result = detector.detect_async(...) line.

I tried to install mediapipe through cmd with python under windows but it doesn’t work