MS COCO Object Detection & Segmentation

Object detection and instance segmentation annotation following the MS COCO format (Lin et al., ECCV 2014). Annotate objects with bounding boxes and polygon segmentation masks across 80 common object categories.

🖼️

image annotation

Configuration Fileconfig.yaml

# MS COCO Object Detection & Segmentation Configuration
# Based on Lin et al., ECCV 2014

annotation_task_name: "MS COCO Object Detection & Segmentation"

data_files:
  - "sample-data.json"

item_properties:
  id_key: "id"
  text_key: "image_url"
  context_key: "context"

user_config:
  allow_all_users: true

annotation_schemes:
  - annotation_type: "multiselect"
    name: "object_categories"
    description: "Select all object categories visible in the image"
    labels:
      - name: "person"
        tooltip: "Human figures, any age"
      - name: "bicycle"
        tooltip: "Bicycles, including parts"
      - name: "car"
        tooltip: "Cars, sedans, coupes"
      - name: "motorcycle"
        tooltip: "Motorcycles, scooters"
      - name: "airplane"
        tooltip: "Aircraft of any type"
      - name: "bus"
        tooltip: "Buses, shuttles"
      - name: "train"
        tooltip: "Trains, trams, metros"
      - name: "truck"
        tooltip: "Trucks, vans, pickups"
      - name: "boat"
        tooltip: "Boats, ships, watercraft"
      - name: "dog"
        tooltip: "Dogs of any breed"
      - name: "cat"
        tooltip: "Cats, domestic felines"
      - name: "horse"
        tooltip: "Horses, ponies"
      - name: "chair"
        tooltip: "Chairs, seats"
      - name: "couch"
        tooltip: "Sofas, couches"
      - name: "dining_table"
        tooltip: "Tables for dining"
      - name: "tv"
        tooltip: "Television sets, monitors"
      - name: "laptop"
        tooltip: "Laptop computers"
      - name: "cell_phone"
        tooltip: "Mobile phones"
      - name: "bottle"
        tooltip: "Bottles of any kind"
      - name: "cup"
        tooltip: "Cups, mugs, glasses"

  - annotation_type: "text"
    name: "bounding_boxes"
    description: "Draw bounding boxes around each object (format: category,x,y,width,height per line)"

  - annotation_type: "radio"
    name: "image_quality"
    description: "Rate the overall image quality for annotation"
    labels:
      - name: "excellent"
        tooltip: "Clear, well-lit, easy to annotate"
      - name: "good"
        tooltip: "Minor issues but annotatable"
      - name: "poor"
        tooltip: "Difficult to annotate due to quality"
      - name: "unusable"
        tooltip: "Cannot be reliably annotated"

interface_config:
  item_display_format: "<img src='{{text}}' style='max-width:100%; max-height:500px;'/><br/><small>{{context}}</small>"

output_annotation_format: "json"
output_annotation_dir: "annotations"

Sample Datasample-data.json

[
  {
    "id": "coco_001",
    "image_url": "https://images.cocodataset.org/val2017/000000397133.jpg",
    "context": "Street scene with various objects. Identify and mark all visible objects."
  },
  {
    "id": "coco_002",
    "image_url": "https://images.cocodataset.org/val2017/000000037777.jpg",
    "context": "Indoor scene. Identify all objects and their locations."
  }
]

// ... and 1 more items

Get This Design

View on GitHub

Clone or download from the repository

Quick start:

git clone https://github.com/davidjurgens/potato-showcase.git
cd potato-showcase/ms-coco
potato start config.yaml

Details

Annotation Types

bboxpolygon

Domain

Computer VisionObject Detection

Use Cases

Object DetectionInstance SegmentationScene Understanding

Related Designs

BDD100K Autonomous Driving Segmentation

Large-scale diverse driving video dataset (Yu et al., CVPR 2020). Annotate driving scenes with bounding boxes, lane markings, drivable areas, and full-frame instance segmentation.

bboxpolygon

MVTec AD Industrial Defect Detection

Anomaly detection and localization in industrial images (Bergmann et al., CVPR 2019). Detect defects across 15 object and texture categories including metal nuts, transistors, and leather.

radiobbox

Open Images V6 Object Detection

Large-scale object detection following Open Images V6 (Kuznetsova et al., IJCV 2020). Annotate 600 object classes with bounding boxes, visual relationships, and instance segmentation masks.

bboxmultiselect