Door Detection & Classification Dataset

This dataset is designed for door object detection, classification, and semantic segmentation. Images were captured with a 3D Realsense D435 camera, with the angle adjusted to ensure the complete door region. The dataset is divided into two subsets: one containing 240 images for door detection/semantic segmentation, and another containing 1,206 images for door classification. The classification categories are closed doors, partially open doors, and fully open doors, and the dataset provides detailed training, validation, and test splits.

Updated 4/2/2024

github

Description

Dataset Overview

Dataset Name

[Door Detection & Classification] Dataset

Dataset Purpose

For detection, classification, and semantic segmentation of door objects.

Data Acquisition Equipment

Using a 3D Realsense camera, model D435, rotated 90° to accommodate the horizontal view of doors.

Data Source

Images captured from Universidade da Beira Interior, public spaces, and private residences, containing doors of various textures and sizes along with surrounding environments.

Sub‑Dataset Details

1. Door Detection | Semantic Segmentation SubDataset

Number of Images: 240
Image Size: 480 x 640 pixels
Image Type: Grayscale
Annotation Information: Pixel value 1 indicates door or door frame, 2 indicates non‑door area
Additional Information: RGB version of annotation images provided for visualization

2. Door Classification SubDataset [3D and RGB]

Number of Images: 1,206 RGB images with corresponding 1,206 depth images
Image Size: 480 x 640 pixels
Depth Image Information: Grayscale image, pixel value range 0‑255, depth unit is 1/16 meter
Door State Categories: 468 images of closed doors, 588 images of open doors, 150 images of semi‑open doors
Dataset Split: Divided into training, validation, and test sets; validation and test each contain 60 images, training set contains 1,086 images
Image Processing: Available in original size and cropped version; the cropped version contains only the door and doorframe region

Dataset Structure

Door Detection
- Annotations
  - Test
  - Train
- Annotations-visualization
  - Test
  - Train
- Images
  - Test
  - Train
Door Classification
- Cropped
  - Depth
    - Test
      - Closed
      - Open
      - Semi-open
    - Train
      - Closed
      - Open
      - Semi-open
    - Val
      - Closed
      - Open
      - Semi-open
  - RGB
    - Test
      - Closed
      - Open
      - Semi-open
    - Train
      - Closed
      - Open
      - Semi-open
    - Val
      - Closed
      - Open
      - Semi-open
- OriginalSize
  - Depth
    - Test
      - Closed
      - Open
      - Semi-open
    - Train
      - Closed
      - Open
      - Semi-open
    - Val
      - Closed
      - Open
      - Semi-open
  - RGB
    - Test
      - Closed
      - Open
      - Semi-open
    - Train
      - Closed
      - Open
      - Semi-open
    - Val
      - Closed
      - Open
      - Semi-open

AI studio

Generate PPTs instantly with Nano Banana Pro.

Generate PPT Now

Access Dataset

Please login to view download links and access full dataset details.

Topics

Computer Vision

Image Classification

Source

Organization: github

Created: 3/26/2020

Power Your Data Analysis with Premium AI Models

Supporting GPT-5, Claude-4, DeepSeek v3, Gemini and more.

Enjoy a free trial and save 20%+ compared to official pricing.

Check Prices →