{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "c4b47595-ee12-4217-b52f-0ec867f74d95",
   "metadata": {},
   "source": [
    "# Visualizing 3D Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "19814cbe-7df1-4129-b77b-16dbd6426122",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "os.environ[\"WEBRTC_PORT\"] = \"8889\"\n",
    "os.environ[\"WEBRTC_IP\"] = \"127.0.0.1\"\n",
    "\n",
    "import pickle\n",
    "from vis4d.vis.functional import show_points\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fac0fb90-4121-4a7a-8a53-9af164ce1500",
   "metadata": {},
   "source": [
    "First, lets load the pointcloud data so we can visualize it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "05d6da4e-ae42-465d-b781-76c82617cb47",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded pointcloud with 535899 points\n"
     ]
    }
   ],
   "source": [
    "pc_data = pickle.load(open(\"data/pc_data.pkl\", \"rb\"))\n",
    "\n",
    "xyz = pc_data[\"points3d\"] - np.mean(pc_data[\"points3d\"], 0)\n",
    "colors = pc_data[\"colors3d\"]\n",
    "classes = pc_data[\"semantics3d\"]\n",
    "instances = pc_data[\"instances3d\"]\n",
    "\n",
    "# Lets remove the top part of the pointcloud to see it better\n",
    "lower_mask = xyz[:, -1] < 1.4\n",
    "xyz = xyz[lower_mask, :]\n",
    "colors = colors[lower_mask, :]\n",
    "classes = classes[lower_mask]\n",
    "instances = instances[lower_mask]\n",
    "\n",
    "print(f\"Loaded pointcloud with {colors.shape[0]} points\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6513b7c-ef2b-4b9d-8bb0-1f211f7b6956",
   "metadata": {},
   "source": [
    "Now, lets visualize the pointcloud with color!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "df27a38a-0a36-465a-b144-17b317bb373d",
   "metadata": {},
   "outputs": [],
   "source": [
    "show_points(xyz, colors)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b9c6830-8a8d-4ccb-b015-37b6e874c053",
   "metadata": {},
   "source": [
    "Lets look at some other predictions. Class and Instances"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3f6fcca9-27ac-403a-8a7e-637ccdc1c6ea",
   "metadata": {},
   "source": [
    "#### Classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "da9e8553-3c9e-45a5-809e-9146b7151755",
   "metadata": {},
   "outputs": [],
   "source": [
    "show_points(xyz, classes = classes)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c880f26-895f-46df-a55e-b6d71e35b15f",
   "metadata": {},
   "source": [
    "#### Instances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "6dd29007-c296-4e39-8cf7-2c787e9b2e82",
   "metadata": {},
   "outputs": [],
   "source": [
    "show_points(xyz, instances = instances)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "69fda6f2-50af-4e08-a185-c29cbcd2643e",
   "metadata": {},
   "source": [
    "Finally, lets show all predictions in one window. \n",
    "This will copy the room three times and color it differently each time, based on 'color', 'classes' and 'instances'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "b2b2a96e-ddda-42ae-bcd3-71f187ea9378",
   "metadata": {},
   "outputs": [],
   "source": [
    "show_points(xyz, colors =colors, classes=classes, instances = instances)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}