Sign Language Detection using ACTION RECOGNITION with Python | LSTM Deep Learning Model
Why Stack Overflow recommends it
Stack Overflow users cite it as the tutorial they were following when they hit errors with Keras imports, OpenCV webcam capture on M1 Macs, and MediaPipe version changes.
Recommended by 0 people across 0 different months, 2021 to 2023.
What it is
A 2.5-hour hands-on tutorial by Nicholas Renotte on building a sign language recognition system in Python. It extracts MediaPipe Holistic keypoints, collects sequences of them, and trains an LSTM network with TensorFlow and Keras to classify actions, ending with real-time testing through a webcam.
What people said
Cites teeara · 2021-10 · Stack Overflow · on “Opencv feed keeps loading after use, jupyter notebook kernel died. I am guessing that the opencv has a problem but I don't know how to solve it”
Python quit unexpectedly. The kernel appears to have died. It will restart automatically. [LINK] (Device: MacBook Air M1) I was collecting keypoints for training an action detection program ( [this video] ) : cap = cv2.VideoCapture(0) Set mediapipe model with mp_holistic.Holistic(min_detection_confidence=0.5, min_tracking_confidence=0.5) as holistic: for action in actions: # Loop through sequences aka videos for sequence in range(no_sequences): # Loop through video length aka sequence length, each frame, 30 frames of keypoints for each video for frame_num in range(sequence_length): # Read feed ret, frame = cap.read() # Make detections image, results = mediapipe_detection(frame, holistic) print th
Cites user20008399 · 2022-09 · Stack Overflow · on “Trying to run "from tensorflow.keras.utils import to_categorical" in jupyter notebooks on mac m1”
I was doing this tutorial: Sign Language Detection using ACTION RECOGNITION with Python [this video] And I got all the way to Part 6: Preprocess Data and Create Labels and Features The issue is this line of code: from tensorflow.keras.utils import to_categorical Giving me these errors: --------------------------------------------------------------------------- TypeError Traceback (most recent call last) Input In [56], in <cell line: 2>() 1 from sklearn.model_selection import train_test_split ----> 2 from tensorflow.keras.utils import to_categorical File /opt/anaconda3/lib/python3.9/site-packages/tensorflow/__init__.py:37, in <module> 35 del estimator_api 36 except (ImportError, AttributeError): -
Cites JJJohn · 2023-06 · Stack Overflow · on “Where can I find the generated binaries when building MediaPipe from source?”
@BenjaminPeterson Because a MediaPipe tutorial [this video] uses MediaPipe 0.8.5 and the latest version doesn't support `mediapipe.solutions.holistic.FACE_CONNECTIONS`