main.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "from tensorflow.keras.layers import Dense, Flatten, Conv2D, MaxPooling2D, Dropout, Bidirectional, LSTM, Reshape, Input\n",
    "from tensorflow.keras.models import Model, Sequential\n",
    "from tensorflow.keras.utils import image_dataset_from_directory \n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.losses import CategoricalCrossentropy\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "physical_devices = tf.config.list_physical_devices('GPU')\n",
    "tf.config.experimental.set_virtual_device_configuration(\n",
    "    physical_devices[0],\n",
    "    [tf.config.experimental.VirtualDeviceConfiguration(memory_limit=6000)]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 2515 files belonging to 36 classes.\n"
     ]
    }
   ],
   "source": [
    "train_dataset = image_dataset_from_directory(\n",
    "    directory='../asl_dataset/',\n",
    "    labels='inferred',\n",
    "    label_mode='categorical',\n",
    "    batch_size=32,\n",
    "    image_size=(256, 256),\n",
    "    shuffle=True,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import cv2\n",
    "import mediapipe as mp\n",
    "import numpy as np\n",
    "mp_hands = mp.solutions.hands\n",
    "hands = mp_hands.Hands(static_image_mode=False, max_num_hands=1)\n",
    "\n",
    "def get_hand_landmarks(image):\n",
    "    results = hands.process(image)\n",
    "\n",
    "    if results.multi_hand_landmarks:\n",
    "        hand_landmarks = results.multi_hand_landmarks[0]\n",
    "\n",
    "        coordinates_3d = np.zeros((21, 3))\n",
    "\n",
    "        for i, landmark in enumerate(hand_landmarks.landmark):\n",
    "            coordinates_3d[i] = [landmark.x, landmark.y, landmark.z]\n",
    "\n",
    "        return np.reshape(coordinates_3d, (63,))\n",
    "    else:\n",
    "        return None\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(63,)\n",
      "(63,)\n",
      "(63,)\n",
      "(63,)\n",
      "(63,)\n",
      "(63,)\n",
      "(63,)\n",
      "(63,)\n",
      "(63,)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "for image_batch, labels_batch in train_dataset:\n",
    "    for i in range(9):\n",
    "        ax = plt.subplot(3, 3, i + 1)\n",
    "        plt.imshow(image_batch[i].numpy().astype(\"uint8\"))\n",
    "        landmark = get_hand_landmarks(image_batch[i].numpy().astype(\"uint8\"))\n",
    "        if not landmark is None:\n",
    "            print(landmark.shape)\n",
    "        # print(type(get_hand_landmarks(image_batch[i].numpy().astype(\"uint8\"))))\n",
    "        plt.axis(\"off\")\n",
    "    break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def landmarks_for_batch(image_batch, labels_batch):\n",
    "    landmarks = []\n",
    "    labels = []\n",
    "    for image,label in zip(image_batch, labels_batch):\n",
    "        landmark = get_hand_landmarks(image.numpy().astype(\"uint8\"))\n",
    "        if not landmark is None:\n",
    "            landmarks.append(landmark)\n",
    "            labels.append(label)\n",
    "    landmarks = np.array(landmarks)\n",
    "    # print(landmarks.shape)\n",
    "    return landmarks, labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(32, 63)\n"
     ]
    }
   ],
   "source": [
    "for image_batch, labels_batch in train_dataset:\n",
    "    landmarks = landmarks_for_batch(image_batch)\n",
    "    print(landmarks.shape)\n",
    "    break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "36"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "class_labels = train_dataset.class_names\n",
    "len(class_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " input_1 (InputLayer)        [(None, 63)]              0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 128)               8192      \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 128)               16512     \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 36)                4644      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 29,348\n",
      "Trainable params: 29,348\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model_inputs = Input(shape=(63,))\n",
    "x = Dense(128, activation='relu')(model_inputs)\n",
    "x = Dense(128, activation='relu')(x)\n",
    "\n",
    "model_outputs = Dense(len(class_labels), activation='softmax')(x)\n",
    "\n",
    "model = Model(model_inputs, model_outputs)\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "OPTIMIZER = Adam(learning_rate=0.001)\n",
    "LOSS = CategoricalCrossentropy()\n",
    "EPOCHS = 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1 Loss: 1.598523736000061\n",
      "Epoch 2 Loss: 1.2799630165100098\n",
      "Epoch 3 Loss: 1.1379497051239014\n",
      "Epoch 4 Loss: 0.812238872051239\n",
      "Epoch 5 Loss: 0.712739884853363\n"
     ]
    }
   ],
   "source": [
    "def train_block(model, train_dataset, epochs):\n",
    "    for i in range(1, epochs + 1):\n",
    "\n",
    "        for image_batch, labels_batch in train_dataset:\n",
    "            with tf.GradientTape() as tape:\n",
    "                landmarks, labels = landmarks_for_batch(image_batch, labels_batch)\n",
    "                # print(landmarks.shape)\n",
    "                preds = model(landmarks)\n",
    "                # print(preds.shape)\n",
    "                loss = LOSS(labels, preds)\n",
    "            gradients = tape.gradient(loss, model.trainable_variables)\n",
    "            OPTIMIZER.apply_gradients(zip(gradients, model.trainable_variables))\n",
    "        print(f'Epoch {i} Loss: {loss.numpy()}')\n",
    "train_block(model, train_dataset, EPOCHS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n"
     ]
    }
   ],
   "source": [
    "model.save('model.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 23ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 37ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 26ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 33ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 29ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 41ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 27ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 16ms/step\n",
      "d\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "5\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "l\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "3\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "3\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "3\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "3\n",
      "1/1 [==============================] - 0s 17ms/step\n",
      "3\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "3\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "3\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "3\n",
      "1/1 [==============================] - 0s 20ms/step\n",
      "3\n"
     ]
    }
   ],
   "source": [
    "import cv2\n",
    "import mediapipe as mp\n",
    "\n",
    "mp_hands = mp.solutions.hands\n",
    "hands = mp_hands.Hands()\n",
    "cap = cv2.VideoCapture(0)\n",
    "mp_drawing = mp.solutions.drawing_utils\n",
    "while True:\n",
    "    ret, frame = cap.read()\n",
    "    frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)\n",
    "    results = hands.process(frame)\n",
    "    landmarks = get_hand_landmarks(frame)\n",
    "    if not landmarks is None:\n",
    "        preds = model.predict(np.expand_dims(landmarks, axis=0))\n",
    "        class_label = class_labels[np.argmax(preds)]\n",
    "        print(class_label)\n",
    "    if results.multi_hand_landmarks:\n",
    "        for hand_landmarks in results.multi_hand_landmarks:\n",
    "            mp_drawing.draw_landmarks(frame, hand_landmarks, mp_hands.HAND_CONNECTIONS)\n",
    "    frame = cv2.cvtColor(frame, cv2.COLOR_RGB2BGR)\n",
    "    cv2.imshow('Hand Tracking', frame)\n",
    "    if cv2.waitKey(1) & 0xFF == ord('q'):\n",
    "        break\n",
    "\n",
    "cap.release()\n",
    "cv2.destroyAllWindows()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "tf",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.18"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}