{ "cells": [ { "cell_type": "markdown", "metadata": { "id": "ZjkBUWm8ZMlc" }, "source": [ "##### Copyright 2026 Google LLC." ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "cellView": "form", "id": "bOTfaUaSZKfF" }, "outputs": [], "source": [ "# @title Licensed under the Apache License, Version 2.0 (the \"License\");\n", "# you may not use this file except in compliance with the License.\n", "# You may obtain a copy of the License at\n", "#\n", "# https://www.apache.org/licenses/LICENSE-2.0\n", "#\n", "# Unless required by applicable law or agreed to in writing, software\n", "# distributed under the License is distributed on an \"AS IS\" BASIS,\n", "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n", "# See the License for the specific language governing permissions and\n", "# limitations under the License." ] }, { "cell_type": "markdown", "metadata": { "id": "CZC64QGBZ2v9" }, "source": [ "# Gemini Quickstart: Gemini Robotics-ER 1.5" ] }, { "cell_type": "markdown", "metadata": { "id": "jMOdCB5FRVMZ" }, "source": [ "" ] }, { "cell_type": "markdown", "metadata": { "id": "eDDzTW8DRVcc" }, "source": [ "This notebook introduces the **Gemini Robotics-ER 1.5** model.\n", "\n", "Gemini Robotics-ER 1.5 is a vision-language model (VLM) that brings Gemini's agentic capabilities to robotics. It's designed for advanced reasoning in the physical world, allowing robots to interpret complex visual data, perform spatial reasoning, and plan actions from natural language commands.\n", "\n", "Key features and benefits:\n", "\n", "* **Enhanced autonomy:** Robots can reason, adapt, and respond to changes in open-ended environments.\n", "* **Natural language interaction:** Makes robots easier to use by enabling complex task assignments using natural language.\n", "* **Task orchestration:** Deconstructs natural language commands into subtasks and integrates with existing robot controllers and behaviors to complete long-horizon tasks.\n", "* **Versatile capabilities:** Locates and identifies objects, understands object relationships, plans grasps and trajectories, and interprets dynamic scenes." ] }, { "cell_type": "markdown", "metadata": { "id": "81jqmyhKZTZ1" }, "source": [ "## Setup\n", "\n", "This section will need to be run any time you start up Colab. The example sections following this are intended to be able to be run without reliance on any other example section, so you may skip through to the examples that are most relevant/interesting to you at the time (though it is strongly recommended that you read through each of these examples at least once!)" ] }, { "cell_type": "markdown", "metadata": { "id": "0cdSPBCltkeQ" }, "source": [ "### Install SDK" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "id": "ONn2UcU7akFl" }, "outputs": [], "source": [ "%pip install -U -q google-genai" ] }, { "cell_type": "markdown", "metadata": { "id": "PD3FmivAUXPk" }, "source": [ "### Setup your API key\n", "\n", "To run the following cells, your API key must be stored it in a Colab Secret named `GOOGLE_API_KEY`. If you don't already have an API key, or you're not sure how to create a Colab Secret, see [Authentication](https://github.com/google-gemini/cookbook/blob/main/quickstarts/Authentication.ipynb) for an example." ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "id": "6L0iD346and2" }, "outputs": [], "source": [ "from google.colab import userdata\n", "\n", "GOOGLE_API_KEY = userdata.get(\"GOOGLE_API_KEY\")" ] }, { "cell_type": "markdown", "metadata": { "id": "Al7zVIf2UqqY" }, "source": [ "### Initialize SDK client\n", "\n", "Initialize a Gemini SDK client with your API key." ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "id": "1ezNXUK8arKa" }, "outputs": [], "source": [ "from google import genai\n", "from google.genai import types\n", "\n", "client = genai.Client(api_key=GOOGLE_API_KEY)" ] }, { "cell_type": "markdown", "metadata": { "id": "95E3SYATU_u4" }, "source": [ "### Select the Gemini Robotics-ER 1.5 model and test\n" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "id": "wIcki0r8VNHY" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Yes, I am here. What can I help you with?\n" ] } ], "source": [ "MODEL_ID = \"gemini-robotics-er-1.5-preview\"\n", "\n", "print(\n", " client.models.generate_content(\n", " model=MODEL_ID, contents=\"Are you there?\"\n", " ).text\n", ")" ] }, { "cell_type": "markdown", "metadata": { "id": "U-70SggjWURa" }, "source": [ "### Imports and utility code" ] }, { "cell_type": "markdown", "metadata": { "id": "E12V7NJ6WlMV" }, "source": [ "#### Imports" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "id": "P37Dv6YQaYw9" }, "outputs": [], "source": [ "import json\n", "import textwrap\n", "import time" ] }, { "cell_type": "markdown", "metadata": { "id": "MgFZ72Bva1FU" }, "source": [ "#### Parsing JSON output" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "id": "ZH6qyE1Uacoq" }, "outputs": [], "source": [ "def parse_json(json_output):\n", " # Parsing out the markdown fencing\n", " lines = json_output.splitlines()\n", " for i, line in enumerate(lines):\n", " if line == \"```json\":\n", " # Remove everything before \"```json\"\n", " json_output = \"\\n\".join(lines[i + 1 :])\n", " # Remove everything after the closing \"```\"\n", " json_output = json_output.split(\"```\")[0]\n", " break # Exit the loop once \"```json\" is found\n", " return json_output" ] }, { "cell_type": "markdown", "metadata": { "id": "XRlsn7tJYUjI" }, "source": [ "#### Resize images\n", "\n", "Resize images for faster rendering and smaller API calls." ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "id": "RMbTihD9YU9D" }, "outputs": [], "source": [ "from PIL import Image\n", "\n", "def get_image_resized(img_path):\n", " img = Image.open(img_path)\n", " img = img.resize(\n", " (800, int(800 * img.size[1] / img.size[0])), Image.Resampling.LANCZOS\n", " )\n", " return img" ] }, { "cell_type": "markdown", "metadata": { "id": "VuvfFW2sa1wu" }, "source": [ "#### Visualization helpers" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "id": "yQQ36nsiX0w-" }, "outputs": [], "source": [ "import base64\n", "import dataclasses\n", "from io import BytesIO\n", "import numpy as np\n", "from PIL import ImageColor, ImageDraw, ImageFont\n", "from typing import Tuple\n", "\n", "import IPython\n", "from IPython import display\n", "\n", "def generate_point_html(pil_image, points_json):\n", " buffered = BytesIO()\n", " pil_image.save(buffered, format=\"PNG\")\n", " img_str = base64.b64encode(buffered.getvalue()).decode()\n", " points_json = parse_json(points_json)\n", "\n", " return f\"\"\"\n", "\n", "\n", "\n", " Point Visualization\n", " \n", "\n", "\n", "
\n", " \n", "
\n", "
\n", "\n", " \n", "\n", "\n", "\"\"\"\n", "\n", "\n", "\n", "additional_colors = [\n", " colorname for (colorname, colorcode) in ImageColor.colormap.items()\n", "]\n", "\n", "\n", "def plot_bounding_boxes(img, bounding_boxes):\n", " \"\"\"Plots bounding boxes on an image.\n", "\n", " Plots bounding boxes on an image with markers for each a name, using PIL,\n", " normalized coordinates, and different colors.\n", "\n", " Args:\n", " img_path: The path to the image file.\n", " bounding_boxes: A list of bounding boxes containing the name of the object\n", " and their positions in normalized [y1 x1 y2 x2] format.\n", " \"\"\"\n", "\n", " # Load the image\n", " width, height = img.size\n", " print(img.size)\n", " # Create a drawing object\n", " draw = ImageDraw.Draw(img)\n", "\n", " # Define a list of colors\n", " colors = [\n", " \"red\",\n", " \"green\",\n", " \"blue\",\n", " \"yellow\",\n", " \"orange\",\n", " \"pink\",\n", " \"purple\",\n", " \"brown\",\n", " \"gray\",\n", " \"beige\",\n", " \"turquoise\",\n", " \"cyan\",\n", " \"magenta\",\n", " \"lime\",\n", " \"navy\",\n", " \"maroon\",\n", " \"teal\",\n", " \"olive\",\n", " \"coral\",\n", " \"lavender\",\n", " \"violet\",\n", " \"gold\",\n", " \"silver\",\n", " ] + additional_colors\n", "\n", " # Parsing out the markdown fencing\n", " bounding_boxes = parse_json(bounding_boxes)\n", "\n", " font = ImageFont.truetype(\"LiberationSans-Regular.ttf\", size=14)\n", "\n", " # Iterate over the bounding boxes\n", " for i, bounding_box in enumerate(json.loads(bounding_boxes)):\n", " # Select a color from the list\n", " color = colors[i % len(colors)]\n", "\n", " # Convert normalized coordinates to absolute coordinates\n", " abs_y1 = int(bounding_box[\"box_2d\"][0] / 1000 * height)\n", " abs_x1 = int(bounding_box[\"box_2d\"][1] / 1000 * width)\n", " abs_y2 = int(bounding_box[\"box_2d\"][2] / 1000 * height)\n", " abs_x2 = int(bounding_box[\"box_2d\"][3] / 1000 * width)\n", "\n", " if abs_x1 > abs_x2:\n", " abs_x1, abs_x2 = abs_x2, abs_x1\n", "\n", " if abs_y1 > abs_y2:\n", " abs_y1, abs_y2 = abs_y2, abs_y1\n", "\n", " # Draw the bounding box\n", " draw.rectangle(((abs_x1, abs_y1), (abs_x2, abs_y2)), outline=color, width=4)\n", "\n", " # Draw the text\n", " if \"label\" in bounding_box:\n", " draw.text(\n", " (abs_x1 + 8, abs_y1 + 6), bounding_box[\"label\"], fill=color, font=font\n", " )\n", "\n", " # Display the image\n", " img.show()\n", "\n", "\n", "@dataclasses.dataclass(frozen=True)\n", "class SegmentationMask:\n", " # bounding box pixel coordinates (not normalized)\n", " y0: int # in [0..height - 1]\n", " x0: int # in [0..width - 1]\n", " y1: int # in [0..height - 1]\n", " x1: int # in [0..width - 1]\n", " mask: np.array # [img_height, img_width] with values 0..255\n", " label: str\n", "\n", "\n", "def parse_segmentation_masks(\n", " predicted_str: str, *, img_height: int, img_width: int\n", ") -> list[SegmentationMask]:\n", " items = json.loads(parse_json(predicted_str))\n", " masks = []\n", " for item in items:\n", " raw_box = item[\"box_2d\"]\n", " abs_y0 = int(item[\"box_2d\"][0] / 1000 * img_height)\n", " abs_x0 = int(item[\"box_2d\"][1] / 1000 * img_width)\n", " abs_y1 = int(item[\"box_2d\"][2] / 1000 * img_height)\n", " abs_x1 = int(item[\"box_2d\"][3] / 1000 * img_width)\n", " if abs_y0 >= abs_y1 or abs_x0 >= abs_x1:\n", " print(\"Invalid bounding box\", item[\"box_2d\"])\n", " continue\n", " label = item[\"label\"]\n", " png_str = item[\"mask\"]\n", " if not png_str.startswith(\"data:image/png;base64,\"):\n", " print(\"Invalid mask\")\n", " continue\n", " png_str = png_str.removeprefix(\"data:image/png;base64,\")\n", " png_str = base64.b64decode(png_str)\n", " mask = Image.open(BytesIO(png_str))\n", " bbox_height = abs_y1 - abs_y0\n", " bbox_width = abs_x1 - abs_x0\n", " if bbox_height < 1 or bbox_width < 1:\n", " print(\"Invalid bounding box\")\n", " continue\n", " mask = mask.resize(\n", " (bbox_width, bbox_height), resample=Image.Resampling.BILINEAR\n", " )\n", " np_mask = np.zeros((img_height, img_width), dtype=np.uint8)\n", " np_mask[abs_y0:abs_y1, abs_x0:abs_x1] = mask\n", " masks.append(\n", " SegmentationMask(abs_y0, abs_x0, abs_y1, abs_x1, np_mask, label)\n", " )\n", " return masks\n", "\n", "\n", "def overlay_mask_on_img(\n", " img: Image, mask: np.ndarray, color: str, alpha: float = 0.7\n", ") -> Image.Image:\n", " \"\"\"Overlays a single mask onto a PIL Image using a named color.\n", "\n", " The mask image defines the area to be colored. Non-zero pixels in the\n", " mask image are considered part of the area to overlay.\n", "\n", " Args:\n", " img: The base PIL Image object.\n", " mask: A PIL Image object representing the mask. Should have the same\n", " height and width as the img. Modes '1' (binary) or 'L' (grayscale) are\n", " typical, where non-zero pixels indicate the masked area.\n", " color: A standard color name string (e.g., 'red', 'blue', 'yellow').\n", " alpha: The alpha transparency level for the overlay (0.0 fully\n", " transparent, 1.0 fully opaque). Default is 0.7 (70%).\n", "\n", " Returns:\n", " A new PIL Image object (in RGBA mode) with the mask overlaid.\n", "\n", " Raises:\n", " ValueError: If color name is invalid, mask dimensions mismatch img\n", " dimensions, or alpha is outside the 0.0-1.0 range.\n", " \"\"\"\n", " if not (0.0 <= alpha <= 1.0):\n", " raise ValueError(\"Alpha must be between 0.0 and 1.0\")\n", "\n", " # Convert the color name string to an RGB tuple\n", " try:\n", " color_rgb: Tuple[int, int, int] = ImageColor.getrgb(color)\n", " except ValueError as e:\n", " # Re-raise with a more informative message if color name is invalid\n", " raise ValueError(\n", " f\"Invalid color name '{color}'. Supported names are typically HTML/CSS \"\n", " f\"color names. Error: {e}\"\n", " )\n", "\n", " # Prepare the base image for alpha compositing\n", " img_rgba = img.convert(\"RGBA\")\n", " width, height = img_rgba.size\n", "\n", " # Create the colored overlay layer\n", " # Calculate the RGBA tuple for the overlay color\n", " alpha_int = int(alpha * 255)\n", " overlay_color_rgba = color_rgb + (alpha_int,)\n", "\n", " # Create an RGBA layer (all zeros = transparent black)\n", " colored_mask_layer_np = np.zeros((height, width, 4), dtype=np.uint8)\n", "\n", " # Mask has values between 0 and 255, threshold at 127 to get binary mask.\n", " mask_np_logical = mask > 127\n", "\n", " # Apply the overlay color RGBA tuple where the mask is True\n", " colored_mask_layer_np[mask_np_logical] = overlay_color_rgba\n", "\n", " # Convert the NumPy layer back to a PIL Image\n", " colored_mask_layer_pil = Image.fromarray(colored_mask_layer_np, \"RGBA\")\n", "\n", " # Composite the colored mask layer onto the base image\n", " result_img = Image.alpha_composite(img_rgba, colored_mask_layer_pil)\n", "\n", " return result_img\n", "\n", "\n", "def plot_segmentation_masks(\n", " img: Image, segmentation_masks: list[SegmentationMask]\n", "):\n", " \"\"\"Plots bounding boxes on an image.\n", "\n", " Plots bounding boxes on an image with markers for each a name, using PIL,\n", " normalized coordinates, and different colors.\n", "\n", " Args:\n", " img: The PIL.Image.\n", " segmentation_masks: A string encoding as JSON a list of segmentation masks\n", " containing the name of the object, their positions in normalized [y1 x1\n", " y2 x2] format, and the png encoded segmentation mask.\n", " \"\"\"\n", " # Define a list of colors\n", " colors = [\n", " \"red\",\n", " \"green\",\n", " \"blue\",\n", " \"yellow\",\n", " \"orange\",\n", " \"pink\",\n", " \"purple\",\n", " \"brown\",\n", " \"gray\",\n", " \"beige\",\n", " \"turquoise\",\n", " \"cyan\",\n", " \"magenta\",\n", " \"lime\",\n", " \"navy\",\n", " \"maroon\",\n", " \"teal\",\n", " \"olive\",\n", " \"coral\",\n", " \"lavender\",\n", " \"violet\",\n", " \"gold\",\n", " \"silver\",\n", " ] + additional_colors\n", "\n", " font = ImageFont.load_default()\n", "\n", " # Do this in 3 passes to make sure the boxes and text are always visible.\n", "\n", " # Overlay the mask\n", " for i, mask in enumerate(segmentation_masks):\n", " color = colors[i % len(colors)]\n", " img = overlay_mask_on_img(img, mask.mask, color)\n", "\n", " # Create a drawing object\n", " draw = ImageDraw.Draw(img)\n", "\n", " # Draw the bounding boxes\n", " for i, mask in enumerate(segmentation_masks):\n", " color = colors[i % len(colors)]\n", " draw.rectangle(\n", " ((mask.x0, mask.y0), (mask.x1, mask.y1)), outline=color, width=4\n", " )\n", "\n", " # Draw the text labels\n", " for i, mask in enumerate(segmentation_masks):\n", " color = colors[i % len(colors)]\n", " if mask.label != \"\":\n", " draw.text((mask.x0 + 8, mask.y0 - 20), mask.label, fill=color, font=font)\n", " return img\n", "\n", "\n", "def overlay_points_on_frames(original_frames, points_data_per_frame):\n", " \"\"\"Overlays points on original frames and returns the modified frames.\"\"\"\n", " modified_frames = []\n", "\n", " # Define colors for drawing points (using a consistent color per label for clarity)\n", " label_colors = {}\n", " current_color_index = 0\n", " available_colors = [\n", " \"red\",\n", " \"green\",\n", " \"blue\",\n", " \"yellow\",\n", " \"orange\",\n", " \"pink\",\n", " \"purple\",\n", " \"brown\",\n", " \"gray\",\n", " \"beige\",\n", " \"turquoise\",\n", " \"cyan\",\n", " \"magenta\",\n", " \"lime\",\n", " \"navy\",\n", " \"maroon\",\n", " \"teal\",\n", " \"olive\",\n", " \"coral\",\n", " \"lavender\",\n", " \"violet\",\n", " \"gold\",\n", " \"silver\",\n", " ]\n", "\n", " font = ImageFont.load_default()\n", "\n", " # Check if the number of original frames matches the number of processed data entries\n", " if len(original_frames) != len(points_data_per_frame):\n", " print(\n", " f\"Error: Number of original frames ({len(original_frames)}) does not \"\n", " \"match the number of processed point data entries\"\n", " f\" ({len(points_data_per_frame)}). Cannot overlay points accurately.\"\n", " )\n", " return original_frames # Return original frames if data doesn't match\n", " else:\n", " # Iterate through the frames and draw points\n", " for i, frame_pil in enumerate(original_frames):\n", " # Ensure frame is in RGB mode for drawing\n", " img = frame_pil.convert(\"RGB\")\n", " draw = ImageDraw.Draw(img)\n", " width, height = img.size\n", "\n", " frame_points = points_data_per_frame[i]\n", "\n", " # Draw points on the frame\n", " for point_info in frame_points:\n", " if \"point\" in point_info and \"label\" in point_info:\n", " y_norm, x_norm = point_info[\"point\"]\n", " label = point_info[\"label\"]\n", "\n", " # Get color for the label\n", " if label not in label_colors:\n", " label_colors[label] = available_colors[\n", " current_color_index % len(available_colors)\n", " ]\n", " current_color_index += 1\n", " color = label_colors[label]\n", "\n", " # Convert normalized coordinates to absolute pixel coordinates\n", " abs_x = int(x_norm / 1000.0 * width)\n", " abs_y = int(y_norm / 1000.0 * height)\n", "\n", " # Draw a circle at the point\n", " point_radius = 5\n", " draw.ellipse(\n", " (\n", " abs_x - point_radius,\n", " abs_y - point_radius,\n", " abs_x + point_radius,\n", " abs_y + point_radius,\n", " ),\n", " fill=color,\n", " outline=color,\n", " )\n", "\n", " # Draw the label\n", " # Adjust label position to avoid going out of bounds\n", " label_pos_x = abs_x + point_radius + 2\n", " label_pos_y = (\n", " abs_y - point_radius - 10\n", " if abs_y - point_radius - 10 > 0\n", " else abs_y + point_radius + 2\n", " )\n", " draw.text((label_pos_x, label_pos_y), label, fill=color, font=font)\n", "\n", " # Append the modified PIL Image\n", " modified_frames.append(img)\n", "\n", " print(f\"Processed and drew points on {len(modified_frames)} frames.\")\n", " return modified_frames\n", "\n", "\n", "def display_gif(frames_to_display):\n", " \"\"\"Saves and displays a list of PIL Images as a GIF.\"\"\"\n", " if frames_to_display:\n", " try:\n", " # Save the modified frames as a new GIF\n", " output_gif_path = \"/tmp/annotated_aloha_pen.gif\"\n", " # Duration per frame in milliseconds (adjust as needed, 40ms is 25fps)\n", " duration_ms = 40\n", " # Ensure all frames are in RGB mode before saving as GIF\n", " rgb_frames = [frame.convert(\"RGB\") for frame in frames_to_display]\n", " if rgb_frames:\n", " rgb_frames[0].save(\n", " output_gif_path,\n", " save_all=True,\n", " append_images=rgb_frames[1:],\n", " duration=duration_ms,\n", " loop=0,\n", " )\n", "\n", " # Display the GIF in Colab\n", " display.display(display.Image(output_gif_path))\n", " print(f\"Displayed annotated GIF: {output_gif_path}\")\n", " else:\n", " print(\"No frames to create GIF.\")\n", "\n", " except Exception as e:\n", " print(f\"Error creating or displaying annotated GIF: {e}\")\n", " else:\n", " print(\"No frames to display.\")\n", "\n", "\n", "def extract_frames(gif):\n", " \"\"\"Extracts frames from a GIF and returns a list of PIL Image objects.\"\"\"\n", " frames = []\n", " try:\n", " while True:\n", " # Convert each frame to RGB to ensure compatibility with drawing\n", " frame = gif.convert(\"RGB\")\n", " frames.append(frame)\n", " gif.seek(gif.tell() + 1) # Move to the next frame\n", " except EOFError:\n", " pass # End of sequence\n", "\n", " print(f\"Extracted {len(frames)} frames from the GIF.\")\n", "\n", " return frames\n", "\n", "\n", "def populate_points_for_all_frames(total_frames, step, analyzed_data):\n", " \"\"\"Populates point data for all frames based on analyzed frames.\"\"\"\n", " points_data_all_frames = []\n", " analyzed_data_index = 0\n", " for i in range(total_frames):\n", " if i % step == 0 and analyzed_data_index < len(analyzed_data):\n", " points_data_all_frames.append(analyzed_data[analyzed_data_index])\n", " analyzed_data_index += 1\n", " else:\n", " # For frames that were not analyzed, use the data from the last analyzed\n", " # frame or append an empty list if no frame has been analyzed yet\n", " if analyzed_data_index > 0:\n", " points_data_all_frames.append(analyzed_data[analyzed_data_index - 1])\n", " else:\n", " # Should not happen if frames list is not empty\n", " points_data_all_frames.append([])\n", " return points_data_all_frames" ] }, { "cell_type": "markdown", "metadata": { "id": "pytVkhMqtfnN" }, "source": [ "## Common settings\n", "\n", "Many of the examples work well with thinking turned off to reduce latency. Temperature is set to `0.5` to increase consistency while still allowing some creativity.\n", "\n", "Most of the examples use a common recipe to call the model - an image, a prompt, and a `GenerateContentConfig`. This function reduces code duplication and highlights when different settings are used." ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "id": "EHcUncwNtL-v" }, "outputs": [], "source": [ "def call_gemini_robotics_er(img, prompt, config=None):\n", " default_config = types.GenerateContentConfig(\n", " temperature=0.5,\n", " thinking_config=types.ThinkingConfig(thinking_budget=0)\n", " )\n", "\n", " if config is None:\n", " config = default_config\n", "\n", " image_response = client.models.generate_content(\n", " model=MODEL_ID,\n", " contents=[img, prompt],\n", " config=config,\n", " )\n", "\n", " print(image_response.text)\n", " return parse_json(image_response.text)" ] }, { "cell_type": "markdown", "metadata": { "id": "mPDvMJz6dGhd" }, "source": [ "## 2D Pointing" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "id": "5KBX_F0adKQ5" }, "outputs": [], "source": [ "!wget https://storage.googleapis.com/generativeai-downloads/images/robotics/er-1-5-example-colab/aloha-arms-table.png -O aloha-arms-table.png -q\n", "!wget https://storage.googleapis.com/generativeai-downloads/images/robotics/er-1-5-example-colab/gameboard.png -O gameboard.png -q\n", "!wget https://storage.googleapis.com/generativeai-downloads/images/robotics/er-1-5-example-colab/washers.png -O washer.png -q\n", "!wget https://storage.googleapis.com/generativeai-downloads/images/robotics/er-1-5-example-colab/aloha-pen.gif -O aloha-pen.gif -q" ] }, { "cell_type": "markdown", "metadata": { "id": "cbCHnSG3i_E0" }, "source": [ "### Pointing to Undefined Objects" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "id": "KqFpIQbtdbjE" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "```json\n", "[\n", " {\"point\": [372, 607], \"label\": \"paper bag\"},\n", " {\"point\": [270, 403], \"label\": \"bowl\"},\n", " {\"point\": [338, 301], \"label\": \"bag of peanuts\"},\n", " {\"point\": [503, 362], \"label\": \"bowl with toy food\"},\n", " {\"point\": [617, 467], \"label\": \"measuring spoon\"},\n", " {\"point\": [602, 655], \"label\": \"banana\"},\n", " {\"point\": [702, 567], \"label\": \"lemon\"},\n", " {\"point\": [702, 612], \"label\": \"lime\"},\n", " {\"point\": [757, 487], \"label\": \"green bell pepper toy\"},\n", " {\"point\": [725, 323], \"label\": \"bread slice\"}\n", "]\n", "```\n", "\n", "Total processing time: 2.4208 seconds\n" ] }, { "data": { "text/html": [ "\n", "\n", "\n", "\n", " Point Visualization\n", " \n", "\n", "\n", "
\n", " \n", "
\n", "
\n", "\n", " \n", "\n", "\n" ], "text/plain": [ "" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "img = get_image_resized(\"aloha-arms-table.png\")\n", "\n", "prompt = textwrap.dedent(\"\"\"\\\n", " Point to no more than 10 items in the image. The label returned should be an\n", " identifying name for the object detected.\n", "\n", " The answer should follow the JSON format:\n", " [{\"point\": , \"label\": }, ...]\n", "\n", " The points are in [y, x] format normalized to 0-1000.\"\"\")\n", "\n", "start_time = time.time()\n", "json_output = call_gemini_robotics_er(img, prompt)\n", "\n", "print(f\"\\nTotal processing time: {(time.time() - start_time):.4f} seconds\")\n", "IPython.display.HTML(generate_point_html(img, json_output))" ] }, { "cell_type": "markdown", "metadata": { "id": "LnSGgHMyjEx4" }, "source": [ "### Pointing to Defined Objects" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "id": "88XlG1f8-uH_" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "```json\n", "[\n", " {\"point\": [759, 317], \"label\": \"bread\"},\n", " {\"point\": [720, 246], \"label\": \"bread\"},\n", " {\"point\": [652, 303], \"label\": \"bread\"},\n", " {\"point\": [837, 792], \"label\": \"bread\"},\n", " {\"point\": [763, 484], \"label\": \"starfruit\"},\n", " {\"point\": [618, 658], \"label\": \"banana\"}\n", "]\n", "```\n", "\n", "Total processing time: 1.7169 seconds\n" ] }, { "data": { "text/html": [ "\n", "\n", "\n", "\n", " Point Visualization\n", " \n", "\n", "\n", "
\n", " \n", "
\n", "
\n", "\n", " \n", "\n", "\n" ], "text/plain": [ "" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "img = get_image_resized(\"aloha-arms-table.png\")\n", "\n", "queries = [\n", " \"bread\",\n", " \"starfruit\",\n", " \"banana\",\n", "]\n", "\n", "prompt = textwrap.dedent(f\"\"\"\\\n", "Get all points matching the following objects: {', '.join(queries)}. The label\n", "returned should be an identifying name for the object detected.\n", "\n", "The answer should follow the JSON format:\n", "[{{\"point\": , \"label\": }}, ...]\n", "\n", "The points are in [y, x] format normalized to 0-1000.\n", "\"\"\")\n", "\n", "start_time = time.time()\n", "json_output = call_gemini_robotics_er(img, prompt)\n", "\n", "points_data = []\n", "try:\n", " data = json.loads(json_output)\n", " points_data.extend(data)\n", "except json.JSONDecodeError:\n", " print(\"Warning: Invalid JSON response. Skipping.\")\n", "\n", "print(f\"\\nTotal processing time: {(time.time() - start_time):.4f} seconds\")\n", "IPython.display.HTML(generate_point_html(img, json.dumps(points_data)))" ] }, { "cell_type": "markdown", "metadata": { "id": "Mf0mLsgdjPy9" }, "source": [ "### Point to all instances of an object based on more abstract description (e.g. \"fruit\")" ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "id": "3Vdb0xW3djfd" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "```json\n", "[\n", " {\"point\": [519, 368], \"label\": \"fruit\"},\n", " {\"point\": [699, 563], \"label\": \"fruit\"},\n", " {\"point\": [702, 608], \"label\": \"fruit\"},\n", " {\"point\": [624, 663], \"label\": \"fruit\"}\n", "]\n", "```\n", "\n", "Total processing time: 1.7176 seconds\n" ] }, { "data": { "text/html": [ "\n", "\n", "\n", "\n", " Point Visualization\n", " \n", "\n", "\n", "
\n", " \n", "
\n", "
\n", "\n", " \n", "\n", "\n" ], "text/plain": [ "" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "points_data = []\n", "img = get_image_resized(\"aloha-arms-table.png\")\n", "\n", "prompt = textwrap.dedent(f\"\"\"\\\n", " Get all points for fruit. The label returned should be an identifying\n", " name for the object detected.\n", "\n", " The answer should follow the json format:\n", " [{{\"point\": , \"label\": }}, ...]\n", "\n", " The points are in [y, x] format normalized to 0-1000.\"\"\")\n", "\n", "start_time = time.time()\n", "json_output = call_gemini_robotics_er(img, prompt)\n", "\n", "try:\n", " data = json.loads(json_output)\n", " points_data.extend(data)\n", "except json.JSONDecodeError:\n", " print(f\"Warning: Invalid JSON response, skipping.\")\n", "\n", "print(f\"\\nTotal processing time: {(time.time() - start_time):.4f} seconds\")\n", "IPython.display.HTML(generate_point_html(img, json.dumps(points_data)))" ] }, { "cell_type": "markdown", "metadata": { "id": "VXI29BZNjTfM" }, "source": [ "### Point to all instances of an object" ] }, { "cell_type": "code", "execution_count": 15, "metadata": { "id": "SqapTqtbdw3c" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "```json\n", "[\n", " {\"point\": [500, 521], \"label\": \"game board slot\"},\n", " {\"point\": [500, 592], \"label\": \"game board slot\"},\n", " {\"point\": [500, 658], \"label\": \"game board slot\"},\n", " {\"point\": [600, 521], \"label\": \"game board slot\"},\n", " {\"point\": [600, 592], \"label\": \"game board slot\"},\n", " {\"point\": [600, 658], \"label\": \"game board slot\"},\n", " {\"point\": [700, 521], \"label\": \"game board slot\"},\n", " {\"point\": [700, 592], \"label\": \"game board slot\"},\n", " {\"point\": [700, 658], \"label\": \"game board slot\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [569, 422], \"label\": \"X game piece\"},\n", " {\"point\": [649, 340], \"label\": \"X game piece\"},\n", " {\"point\": [471, 788], \"label\": \"X game piece\"},\n", " {\"point\": [549, 836], \"label\": \"X game piece\"}\n", "]\n", "```\n", "\n", "Total processing time: 4.5273 seconds\n" ] }, { "data": { "text/html": [ "\n", "\n", "\n", "\n", " Point Visualization\n", " \n", "\n", "\n", "
\n", " \n", "
\n", "
\n", "\n", " \n", "\n", "\n" ], "text/plain": [ "" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "points_data = []\n", "img = get_image_resized(\"gameboard.png\")\n", "\n", "queries = [\n", " \"game board slot\",\n", " \"X game piece\",\n", "]\n", "\n", "start_time = time.time()\n", "for obj in queries:\n", " prompt = textwrap.dedent(f\"\"\"\\\n", " Get all points matching {obj}. The label returned should be an identifying\n", " name for the object detected.\n", "\n", " The answer should follow the JSON format:\n", " [{{\"point\": , \"label\": }}, ...]\n", "\n", " The points are in [y, x] format normalized to 0-1000.\"\"\")\n", " json_output = call_gemini_robotics_er(img, prompt)\n", "\n", " try:\n", " data = json.loads(json_output)\n", " points_data.extend(data)\n", " except json.JSONDecodeError:\n", " print(f\"Warning: Invalid JSON response for {obj}. Skipping.\")\n", " continue\n", "\n", "print(f\"\\nTotal processing time: {(time.time() - start_time):.4f} seconds\")\n", "IPython.display.HTML(generate_point_html(img, json.dumps(points_data)))" ] }, { "cell_type": "markdown", "metadata": { "id": "y_w8K-ayjWYM" }, "source": [ "### Pointing to certain parts of an object in serial" ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "id": "K3jMpkXndyfa" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "```json\n", "[\n", " {\"point\": [390, 485], \"label\": \"handles\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [564, 616], \"label\": \"the stem of the banana\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [609, 686], \"label\": \"center of the banana\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [763, 484], \"label\": \"starfruit\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [704, 608], \"label\": \"lime\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [308, 417], \"label\": \"rim\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [308, 412], \"label\": \"rim of the dark blue bowl\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [611, 477], \"label\": \"rim\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [619, 410], \"label\": \"handle of the measuring cup\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [521, 368], \"label\": \"tomato\"}\n", "]\n", "```\n", "\n", "Total processing time: 13.2501 seconds\n" ] }, { "data": { "text/html": [ "\n", "\n", "\n", "\n", " Point Visualization\n", " \n", "\n", "\n", "
\n", " \n", "
\n", "
\n", "\n", " \n", "\n", "\n" ], "text/plain": [ "" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "img = get_image_resized(\"aloha-arms-table.png\")\n", "points_data = []\n", "\n", "queries = [\n", " (\"paper bag\", \"handles\"),\n", " (\"banana\", \"the stem\"),\n", " (\"banana\", \"center\"),\n", " (\"starfruit\", \"center\"),\n", " (\"lime\", \"center\"),\n", " (\"light blue bowl\", \"rim\"),\n", " (\"dark blue bowl\", \"rim\"),\n", " (\"measuring cup\", \"rim\"),\n", " (\"measuring cup\", \"handle\"),\n", " (\"bowl\", \"tomato\"),\n", "]\n", "\n", "start_time = time.time()\n", "for obj, part in queries:\n", " POINT_PROMPT_TEMPLATE = textwrap.dedent(\"\"\"\\\n", " Point to the $part of the $object in the image. Return the answer as a\n", " JSON list of a dictionary with keys 'point' and 'label'. Only return one\n", " point for this request.\"\"\")\n", " prompt = POINT_PROMPT_TEMPLATE.replace(\"$object\", obj).replace(\"$part\", part)\n", "\n", " json_output = call_gemini_robotics_er(img, prompt)\n", "\n", " try:\n", " data = json.loads(json_output)\n", " points_data.extend(data)\n", " except json.JSONDecodeError:\n", " print(f\"Warning: Invalid JSON response for {obj}, {part}. Skipping.\")\n", " continue\n", "\n", "print(f\"\\nTotal processing time: {(time.time() - start_time):.4f} seconds\")\n", "IPython.display.HTML(generate_point_html(img, json.dumps(points_data)))" ] }, { "cell_type": "markdown", "metadata": { "id": "thIJjmAKd0yG" }, "source": [ "As you can see in the example above, pointing to these objects in a loop can take a moment (e.g. 15 seconds). When possible, it is recommended that you run your queries in parallel to improve response time, as you can see in the following example." ] }, { "cell_type": "markdown", "metadata": { "id": "oGXj_wrejZnG" }, "source": [ "### Pointing to certain parts of an object in parallel" ] }, { "cell_type": "code", "execution_count": 17, "metadata": { "id": "UmLtTiWSd16y" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "```json\n", "[\n", " {\"point\": [389, 412], \"label\": \"handles\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [609, 688], \"label\": \"center of the banana\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [704, 608], \"label\": \"lime\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [770, 485], \"label\": \"starfruit\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [564, 613], \"label\": \"the stem of the banana\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [232, 401], \"label\": \"rim of the light blue bowl\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [607, 479], \"label\": \"rim of the measuring cup\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [305, 400], \"label\": \"rim of the dark blue bowl\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [619, 411], \"label\": \"handle of the measuring cup\"}\n", "]\n", "```\n", "```json\n", "[\n", " {\"point\": [521, 368], \"label\": \"tomato\"}\n", "]\n", "```\n", "\n", "Total processing time: 4.0269 seconds\n" ] }, { "data": { "text/html": [ "\n", "\n", "\n", "\n", " Point Visualization\n", " \n", "\n", "\n", "
\n", " \n", "
\n", "
\n", "\n", " \n", "\n", "\n" ], "text/plain": [ "" ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import concurrent.futures\n", "\n", "img = get_image_resized(\"aloha-arms-table.png\")\n", "points_data = []\n", "\n", "queries = [\n", " (\"paper bag\", \"handles\"),\n", " (\"banana\", \"the stem\"),\n", " (\"banana\", \"center\"),\n", " (\"starfruit\", \"center\"),\n", " (\"lime\", \"center\"),\n", " (\"light blue bowl\", \"rim\"),\n", " (\"dark blue bowl\", \"rim\"),\n", " (\"measuring cup\", \"rim\"),\n", " (\"measuring cup\", \"handle\"),\n", " (\"bowl\", \"tomato\"),\n", "]\n", "\n", "def process_query(obj, part):\n", " POINT_PROMPT_TEMPLATE = textwrap.dedent(\"\"\"\\\n", " Point to the $part of the $object in the image. Return the answer as a\n", " JSON list of a dictionary with keys 'point' and 'label'. Only return one\n", " point for this request.\"\"\")\n", " prompt = POINT_PROMPT_TEMPLATE.replace(\"$object\", obj).replace(\"$part\", part)\n", " json_output = call_gemini_robotics_er(img, prompt)\n", " try:\n", " data = json.loads(json_output)\n", " return data\n", " except json.JSONDecodeError:\n", " print(f\"Warning: Invalid JSON response for {obj}, {part}. Skipping.\")\n", " return []\n", "\n", "start_time = time.time()\n", "with concurrent.futures.ThreadPoolExecutor() as executor:\n", " results = executor.map(\n", " lambda query: process_query(query[0], query[1]), queries\n", " )\n", "\n", "for result in results:\n", " points_data.extend(result)\n", "\n", "print(f\"\\nTotal processing time: {(time.time() - start_time):.4f} seconds\")\n", "IPython.display.HTML(generate_point_html(img, json.dumps(points_data)))" ] }, { "cell_type": "code", "execution_count": 18, "metadata": { "id": "88zkaMyajc4c" }, "outputs": [], "source": [ "### Counting by Pointing" ] }, { "cell_type": "code", "execution_count": 19, "metadata": { "id": "7O8QCm1Pd3le" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "```json\n", "[\n", " {\"point\": [326, 311], \"label\": \"washer\"},\n", " {\"point\": [320, 401], \"label\": \"washer\"},\n", " {\"point\": [330, 498], \"label\": \"washer\"},\n", " {\"point\": [330, 621], \"label\": \"washer\"},\n", " {\"point\": [372, 651], \"label\": \"washer\"},\n", " {\"point\": [400, 568], \"label\": \"washer\"},\n", " {\"point\": [447, 616], \"label\": \"washer\"},\n", " {\"point\": [440, 483], \"label\": \"washer\"},\n", " {\"point\": [435, 403], \"label\": \"washer\"},\n", " {\"point\": [390, 373], \"label\": \"washer\"},\n", " {\"point\": [386, 296], \"label\": \"washer\"},\n", " {\"point\": [463, 276], \"label\": \"washer\"},\n", " {\"point\": [504, 348], \"label\": \"washer\"},\n", " {\"point\": [549, 260], \"label\": \"washer\"},\n", " {\"point\": [549, 696], \"label\": \"washer\"},\n", " {\"point\": [655, 715], \"label\": \"washer\"}\n", "]\n", "```\n", "count: 16\n", "\n", "Total processing time: 4.5289 seconds\n" ] }, { "data": { "text/html": [ "\n", "\n", "\n", "\n", " Point Visualization\n", " \n", "\n", "\n", "
\n", " \n", "
\n", "
\n", "\n", " \n", "\n", "\n" ], "text/plain": [ "" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "img = get_image_resized(\"washer.png\")\n", "\n", "prompt = textwrap.dedent(\"\"\"\\\n", " Point to each washer in the box. Return the answer in the format:\n", " [{\"point\": , \"label\": }, ...]\n", "\n", " The points are in [y, x] format normalized to 0-1000.\"\"\")\n", "\n", "start_time = time.time()\n", "json_output = call_gemini_robotics_er(img, prompt)\n", "\n", "try:\n", " data = json.loads(json_output)\n", " print(f\"count: {len(data)}\")\n", "except json.JSONDecodeError:\n", " print(\"Error: Could not decode JSON response from the model.\")\n", "\n", "print(f\"\\nTotal processing time: {(time.time() - start_time):.4f} seconds\")\n", "IPython.display.HTML(generate_point_html(img, json_output))" ] }, { "cell_type": "markdown", "metadata": { "id": "qoe0mV5Ujgg8" }, "source": [ "### Pointing to Defined Objects in a GIF" ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "id": "ah7f2SZuYZYi" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Successfully loaded GIF: aloha-pen.gif\n", "Extracted 160 frames from the GIF.\n", "Processing frame 1/160...\n", " Successfully parsed 3 points for frame 1.\n", "Processing frame 11/160...\n", " Successfully parsed 3 points for frame 11.\n", "Processing frame 21/160...\n", " Successfully parsed 3 points for frame 21.\n", "Processing frame 31/160...\n", " Successfully parsed 4 points for frame 31.\n", "Processing frame 41/160...\n", " Successfully parsed 3 points for frame 41.\n", "Processing frame 51/160...\n", " Successfully parsed 3 points for frame 51.\n", "Processing frame 61/160...\n", " Successfully parsed 3 points for frame 61.\n", "Processing frame 71/160...\n", " Successfully parsed 4 points for frame 71.\n", "Processing frame 81/160...\n", " Successfully parsed 4 points for frame 81.\n", "Processing frame 91/160...\n", " Successfully parsed 3 points for frame 91.\n", "Processing frame 101/160...\n", " Successfully parsed 4 points for frame 101.\n", "Processing frame 111/160...\n", " Successfully parsed 3 points for frame 111.\n", "Processing frame 121/160...\n", " Successfully parsed 4 points for frame 121.\n", "Processing frame 131/160...\n", " Successfully parsed 4 points for frame 131.\n", "Processing frame 141/160...\n", " Successfully parsed 3 points for frame 141.\n", "Processing frame 151/160...\n", " Successfully parsed 2 points for frame 151.\n", "Collected point data for 16 analyzed frames.\n", "Populated point data for 160 total frames.\n", "Processed and drew points on 160 frames.\n" ] }, { "data": { "image/jpeg": "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", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Displayed annotated GIF: /tmp/annotated_aloha_pen.gif\n" ] } ], "source": [ "gif_path = \"aloha-pen.gif\"\n", "try:\n", " gif = Image.open(gif_path)\n", " print(f\"Successfully loaded GIF: {gif_path}\")\n", "except FileNotFoundError:\n", " print(f\"Error: GIF file not found at {gif_path}\")\n", " raise\n", "\n", "frames = extract_frames(gif)\n", "\n", "# Define the objects to query\n", "queries = [\n", " \"pen (on desk)\",\n", " \"pen (in robot hand)\",\n", " \"laptop (opened)\",\n", " \"laptop (closed)\",\n", "]\n", "\n", "prompt = textwrap.dedent(f\"\"\"\\\n", "Point to the following objects in the provided image: {\", \".join(queries)}.\n", "\n", "The answer should follow the JSON format:\n", "[{{\"point\": , \"label\": }}, ...]\n", "\n", "The points are in [y, x] format normalized to 0-1000.\n", "\n", "If no objects are found, return an empty JSON list [].\"\"\")\n", "\n", "\n", "# Send every 10th frame as a separate request for the sake of time\n", "analyzed_frames_data = []\n", "frame_step = 10\n", "\n", "for i in range(0, len(frames), frame_step):\n", " frame_index = i\n", " frame = frames[frame_index]\n", " print(f\"Processing frame {frame_index+1}/{len(frames)}...\")\n", "\n", " try:\n", " image_response = client.models.generate_content(\n", " model=MODEL_ID,\n", " contents=[frame, prompt],\n", " config=types.GenerateContentConfig(\n", " temperature=0.5,\n", " thinking_config=types.ThinkingConfig(thinking_budget=0),\n", " ),\n", " )\n", "\n", " try:\n", " json_output = parse_json(image_response.text)\n", " frame_points = json.loads(json_output)\n", " analyzed_frames_data.append(frame_points)\n", " print(\n", " f\" Successfully parsed {len(frame_points)} points for frame\"\n", " f\" {frame_index+1}.\"\n", " )\n", " except json.JSONDecodeError as e:\n", " print(\n", " f\" Error decoding JSON for frame {frame_index+1}: {e}. Appending\"\n", " \" empty list.\"\n", " )\n", " analyzed_frames_data.append([])\n", " except Exception as e:\n", " print(\n", " \" An unexpected error occurred processing frame\"\n", " f\" {frame_index+1} response: {e}. Appending empty list.\"\n", " )\n", " analyzed_frames_data.append([])\n", "\n", " except Exception as e:\n", " print(\n", " f\" Error generating content for frame {frame_index+1}: {e}. Appending\"\n", " \" empty list.\"\n", " )\n", " analyzed_frames_data.append([])\n", "\n", "\n", "print(f\"Collected point data for {len(analyzed_frames_data)} analyzed frames.\")\n", "\n", "points_data_all_frames = populate_points_for_all_frames(\n", " len(frames), frame_step, analyzed_frames_data\n", ")\n", "print(f\"Populated point data for {len(points_data_all_frames)} total frames.\")\n", "\n", "\n", "modified_frames = overlay_points_on_frames(frames, points_data_all_frames)\n", "display_gif(modified_frames)" ] }, { "cell_type": "markdown", "metadata": { "id": "DK9IACHTeETV" }, "source": [ "## Object Detection and Bounding Boxes" ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "id": "eCgpTAPmeG7n" }, "outputs": [], "source": [ "!wget https://storage.googleapis.com/generativeai-downloads/images/robotics/er-1-5-example-colab/aloha-arms-table.png -O aloha-arms-table.png -q" ] }, { "cell_type": "markdown", "metadata": { "id": "RGGCbTnOjkJx" }, "source": [ "### 2D Bounding boxes" ] }, { "cell_type": "code", "execution_count": 25, "metadata": { "id": "2Ekqe-vceI5O" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "```json\n", "[\n", " {\"box_2d\": [734, 725, 952, 865], \"label\": \"bread slice\"},\n", " {\"box_2d\": [544, 534, 614, 579], \"label\": \"green bell pepper toy\"},\n", " {\"box_2d\": [440, 303, 566, 412], \"label\": \"light green bowl\"},\n", " {\"box_2d\": [590, 236, 734, 368], \"label\": \"bread slice\"},\n", " {\"box_2d\": [674, 260, 834, 378], \"label\": \"bread slice\"},\n", " {\"box_2d\": [654, 579, 740, 633], \"label\": \"lime toy\"},\n", " {\"box_2d\": [718, 458, 824, 513], \"label\": \"green apple toy\"},\n", " {\"box_2d\": [646, 203, 784, 264], \"label\": \"bread slice\"},\n", " {\"box_2d\": [544, 608, 704, 780], \"label\": \"banana toy\"},\n", " {\"box_2d\": [496, 340, 560, 398], \"label\": \"tomato toy\"},\n", " {\"box_2d\": [158, 465, 520, 764], \"label\": \"paper bag\"},\n", " {\"box_2d\": [740, 636, 808, 690], \"label\": \"green apple toy\"},\n", " {\"box_2d\": [214, 354, 322, 450], \"label\": \"blue bowl\"},\n", " {\"box_2d\": [674, 538, 762, 592], \"label\": \"lemon toy\"},\n", " {\"box_2d\": [200, 250, 442, 327], \"label\": \"bag of peanuts\"},\n", " {\"box_2d\": [556, 372, 646, 506], \"label\": \"measuring cup\"}\n", "]\n", "```\n", "\n", "Total processing time: 4.6690 seconds\n", "(800, 602)\n" ] }, { "data": { "image/jpeg": "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", "text/plain": [ "" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "img = get_image_resized(\"aloha-arms-table.png\")\n", "\n", "prompt = textwrap.dedent(\"\"\"\\\n", " Return bounding boxes as a JSON array with labels. Never return masks or\n", " code fencing. Limit to 25 objects. Include as many objects as you can\n", " identify on the table.\n", " If an object is present multiple times, name them according to their\n", " unique characteristic (colors, size, position, unique characteristics,\n", " etc..).\n", " The format should be as follows:\n", " [{\"box_2d\": [ymin, xmin, ymax, xmax], \"label\":