{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import warnings\n",
"import os\n",
"warnings.filterwarnings(\"ignore\")\n",
"os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"2024-07-06 13:54:53.658693: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:479] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
"2024-07-06 13:54:53.787338: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:10575] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
"2024-07-06 13:54:53.788034: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1442] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n"
]
}
],
"source": [
"# Pseudocode\n",
"\n",
"# Step 1: Import necessary libraries\n",
"import cv2\n",
"import mediapipe as mp\n",
"import os\n",
"import numpy as np\n",
"from sklearn.neural_network import MLPClassifier\n",
"from sklearn.model_selection import train_test_split\n",
"from sklearn.preprocessing import LabelEncoder\n",
"from tqdm import tqdm"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO: Created TensorFlow Lite XNNPACK delegate for CPU.\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"W0000 00:00:1720254303.391859 3449 inference_feedback_manager.cc:114] Feedback manager requires a model with a single signature inference. Disabling support for feedback tensors.\n",
"W0000 00:00:1720254303.417060 3449 inference_feedback_manager.cc:114] Feedback manager requires a model with a single signature inference. Disabling support for feedback tensors.\n"
]
}
],
"source": [
"\n",
"# Step 2: Initialize MediaPipe Hand model\n",
"mp_hands = mp.solutions.hands\n",
"hands = mp_hands.Hands(static_image_mode=True, max_num_hands=1, min_detection_confidence=0.5)\n",
"\n",
"# Step 3: Function to extract landmarks from an image\n",
"def extract_landmarks(image_path):\n",
" image = cv2.imread(image_path)\n",
" results = hands.process(cv2.cvtColor(image, cv2.COLOR_BGR2RGB))\n",
" if results.multi_hand_landmarks:\n",
" # Extract landmarks\n",
" landmarks = np.array([[lm.x, lm.y, lm.z] for lm in results.multi_hand_landmarks[0].landmark]).flatten()\n",
" return landmarks\n",
" return None\n"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
" \r"
]
}
],
"source": [
"\n",
"# Step 4: Load dataset and extract features\n",
"X = [] # Features\n",
"y = [] # Labels\n",
"\n",
"dataset_path = 'split_asl_dataset'\n",
"for subset in ['train', 'test', 'val']:\n",
" subset_path = os.path.join(dataset_path, subset)\n",
" for label in os.listdir(subset_path):\n",
" class_path = os.path.join(subset_path, label)\n",
" for image_name in tqdm(os.listdir(class_path), desc=f'{label} Images', leave=False):\n",
" image_path = os.path.join(class_path, image_name)\n",
" landmarks = extract_landmarks(image_path)\n",
" if landmarks is not None:\n",
" X.append(landmarks)\n",
" y.append(label)\n"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"'/mnt/d/NullClass_Internship/Action_Detection/Static_Detection'"
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import os\n",
"os.getcwd()"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
"# Convert lists to numpy arrays\n",
"X_np = np.array(X)\n",
"y_np = np.array(y)\n",
"\n",
"# Save to files\n",
"np.save('saved_features/features.npy', X_np)\n",
"np.save('saved_features/labels.npy', y_np)"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
"\n",
"# Step 5: Encode labels\n",
"label_encoder = LabelEncoder()\n",
"y_encoded = label_encoder.fit_transform(y)\n",
"\n",
"# Step 6: Split data into training and testing\n",
"X_train, X_test, y_train, y_test = train_test_split(X, y_encoded, test_size=0.2, random_state=48)\n"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
MLPClassifier(max_iter=500)
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.