Back to Ultralytics

HomeObjects-3K Detection Dataset

docs/en/datasets/detect/homeobjects-3k.md

8.4.918.8 KB
Original Source

HomeObjects-3K Dataset

<a href="https://colab.research.google.com/github/ultralytics/notebooks/blob/main/notebooks/how-to-train-ultralytics-yolo-on-homeobjects-dataset.ipynb"></a>

The Ultralytics HomeObjects-3K dataset is an indoor object detection dataset of 2,689 images (2,285 training and 404 validation) labeled across 12 everyday household classes — bed, sofa, chair, table, lamp, tv, laptop, wardrobe, window, door, potted plant, and photo frame. Built for training and benchmarking computer vision models, it targets indoor scene understanding, smart home devices, robotics, and augmented reality.

<p align="center"> <iframe loading="lazy" width="720" height="405" src="https://www.youtube.com/embed/v3iqOYoRBFQ" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" allowfullscreen> </iframe>

<strong>Watch:</strong> How to Train Ultralytics YOLO on HomeObjects-3K Dataset | Detection, Validation & ONNX Export 🚀

</p>

Dataset Structure

The HomeObjects-3K dataset is split into two predefined subsets, defined by the HomeObjects-3K.yaml configuration:

SplitImagesDescription
Train2,285Indoor scenes with labeled household objects
Validation404Held-out images for evaluating model performance

Each image is labeled with bounding boxes in the Ultralytics YOLO format, ready for object detection and tracking pipelines.

Object Classes

The dataset covers 12 everyday object categories spanning furniture, electronics, and decorative items commonly found in indoor domestic environments:

!!! tip "HomeObjects-3K classes"

0. bed
1. sofa
2. chair
3. table
4. lamp
5. tv
6. laptop
7. wardrobe
8. window
9. door
10. potted plant
11. photo frame

Applications

HomeObjects-3K supports a range of indoor computer vision applications across research and product development:

  • Indoor object detection: Use models like Ultralytics YOLO26 to find and locate common home items like beds, chairs, lamps, and laptops in images for real-time understanding of indoor scenes.

  • Scene layout parsing: Help robotics and smart home systems understand how rooms are arranged — where doors, windows, and furniture sit — so devices can navigate safely and interact with their environment.

  • AR applications: Power object recognition features in augmented reality apps. For example, detect TVs or wardrobes and overlay extra information or effects on them.

  • Education and research: Give students and researchers a ready-to-use dataset for practicing indoor object detection with real-world examples.

  • Home inventory and asset tracking: Automatically detect and list home items in photos or videos, useful for managing belongings, organizing spaces, or visualizing furniture in real estate.

To label your own indoor images, train, and manage dataset versions in your browser, run the full workflow with Ultralytics Platform.

Dataset YAML

The HomeObjects-3K.yaml file defines the dataset configuration — the train and validation image paths and the list of object classes. It is maintained in the Ultralytics repository at https://github.com/ultralytics/ultralytics/blob/main/ultralytics/cfg/datasets/HomeObjects-3K.yaml.

!!! example "ultralytics/cfg/datasets/HomeObjects-3K.yaml"

```yaml
--8<-- "ultralytics/cfg/datasets/HomeObjects-3K.yaml"
```

Usage

You can train a YOLO26n model on the HomeObjects-3K dataset for 100 epochs using an image size of 640. The dataset (390 MB) downloads automatically on first use. The examples below show how to get started. For more training options and detailed settings, check the Training guide.

!!! example "Train Example"

=== "Python"

    ```python
    from ultralytics import YOLO

    # Load pretrained model
    model = YOLO("yolo26n.pt")

    # Train the model on HomeObjects-3K dataset
    model.train(data="HomeObjects-3K.yaml", epochs=100, imgsz=640)
    ```

=== "CLI"

    ```bash
    yolo detect train data=HomeObjects-3K.yaml model=yolo26n.pt epochs=100 imgsz=640
    ```

Sample Images and Annotations

The sample below shows an indoor scene from the dataset with its bounding-box annotations, illustrating the object positions, scales, and spatial relationships that models learn to detect.

License and Attribution

HomeObjects-3K is developed and released by the Ultralytics team under the AGPL-3.0 License, supporting open-source research and commercial use with proper attribution.

If you use this dataset in your research, please cite it using the mentioned details:

!!! quote ""

=== "BibTeX"

    ```bibtex
    @dataset{Jocher_Ultralytics_Datasets_2025,
        author = {Jocher, Glenn and Rizwan, Muhammad},
        license = {AGPL-3.0},
        month = {May},
        title = {Ultralytics Datasets: HomeObjects-3K Detection Dataset},
        url = {https://docs.ultralytics.com/datasets/detect/homeobjects-3k/},
        version = {1.0.0},
        year = {2025}
    }
    ```

FAQ

What is the HomeObjects-3K dataset designed for?

HomeObjects-3K is designed for detecting everyday household items — like beds, sofas, TVs, and lamps — in indoor scenes. This makes it well suited to smart homes, robotics, augmented reality, and interior monitoring systems, for both real-time edge deployment and academic research.

How many images and classes are in the HomeObjects-3K dataset?

HomeObjects-3K contains 2,689 images total — 2,285 for training and 404 for validation — with no separate test split. Every image is labeled across 12 object classes: bed, sofa, chair, table, lamp, tv, laptop, wardrobe, window, door, potted plant, and photo frame.

Which object categories are included, and why were they selected?

The dataset includes 12 of the most commonly encountered household items: bed, sofa, chair, table, lamp, tv, laptop, wardrobe, window, door, potted plant, and photo frame. These objects were chosen to reflect realistic indoor environments and to support tasks such as robotic navigation and scene understanding in AR/VR applications.

How do I download the HomeObjects-3K dataset?

The dataset (390 MB) downloads automatically the first time you train with data="HomeObjects-3K.yaml" — no manual step is required. Ultralytics fetches the images and labels and unpacks them to your local datasets directory. You can browse related datasets in the detection datasets overview.

How can I train a YOLO model using the HomeObjects-3K dataset?

To train a YOLO model like YOLO26n, you need the HomeObjects-3K.yaml configuration file and the pretrained model weights. Training launches with a single Python or CLI command, and you can customize parameters such as epochs, image size, and batch size for your target performance and hardware.

!!! example "Train Example"

=== "Python"

    ```python
    from ultralytics import YOLO

    # Load pretrained model
    model = YOLO("yolo26n.pt")

    # Train the model on HomeObjects-3K dataset
    model.train(data="HomeObjects-3K.yaml", epochs=100, imgsz=640)
    ```

=== "CLI"

    ```bash
    yolo detect train data=HomeObjects-3K.yaml model=yolo26n.pt epochs=100 imgsz=640
    ```

Is this dataset suitable for beginner-level projects?

Yes. Its standardized YOLO-format annotations and compact size make HomeObjects-3K a strong entry point for students and hobbyists exploring real-world object detection in indoor scenarios.

Where can I find the annotation format and YAML?

Refer to the Dataset YAML section. The format is standard YOLO, making it compatible with most object detection pipelines.