---
title: "3D Residential Scans Dataset"
description: "This dataset contains 3D scans of apartments, including splats and meshes generated via 3DGS technology, with high-quality point clouds, lidar scans, RGB-D data, and textured…"
url: "https://unidata.pro/datasets/3d-residential-scans/"
date_modified: "2026-06-17T11:02:43+03:00"
language: "en-US"
---
This dataset contains 3D scans of apartments, including splats and meshes generated via 3DGS technology, with high-quality point clouds, lidar scans, RGB-D data, and textured 3D meshes captured using the XGRIDS PortalCam. Designed for robotics, physical AI, and robot training, this real-world 3DGS dataset supports 3D reconstruction, scene understanding, object detection, spatial analysis, and deep learning workflows for indoor environments.

## List of Parameters

- **Title:** Tasks — **Description:** 3D Reconstruction, Scene Understanding, Mesh Generation
- **Title:** Scan format — **Description:** PLY / LAS (splats), OBJ/GLTF (meshes)
- **Title:** Capture device — **Description:** XGRIDS PortalCam
- **Title:** Number of scans — **Description:** 100
- **Title:** Sensor outputs — **Description:** 3-axis accelerometer, 3-axis gyroscope, 3-axis magnetometer, Orientation quaternions

## Dataset Structure

### The Numbers Section

**Numbered list:** - **Number:** 100 — **Text:** Scans

### Tooltips Section

**Tooltip items:**

- **Name:** Computer Vision
- **Name:** 3D Reconstruction
- **Name:** Mesh Generation
- **Name:** Gaussian Splatting
- **Name:** Scene Understanding

### Dataset Information

**Table with data:**

| Characteristic | Data |
| --- | --- |
| Description | 3D scans of apartments, including splats and meshes generated via 3DGS technology |
| Data types | 3D point clouds (splats), meshes, metadata |
| Tasks | 3D Reconstruction, Scene Understanding, Mesh Generation |
| Number of scans | 100 |
| Accessibility levels | Multiple levels of free movement accessibility |
| Sensor outputs | 3-axis accelerometer, 3-axis gyroscope, 3-axis magnetometer, Orientation quaternions |

**Media Slider:** - **Video on Slayder:** <https://unidata.pro/wp-content/uploads/2026/05/residential-scans.mp4>

**Link to the sample:** [Download sample](https://drive.google.com/drive/u/4/folders/1D8ymeNp0Z97DshDDxC3zY8AtTo2VRhak)

### Technical Specifications

**Table with data:**

| Characteristic | Data |
| --- | --- |
| Scan format | PLY / LAS (splats), OBJ/GLTF (meshes) |
| Capture device | XGRIDS PortalCam |
| Output formats | Splats, Meshes |

**Source and data collection methodology:** Source and collection methodology Data was collected by the Unidata team using the XGRIDS PortalCam device

### Statistics - Charts

**Charts with Titles:** - **Shortcode:** [ays_chart id="28"] — **caption above the graph:** Apartment room distribution

### Dataset Use Cases - Slider

**Industry Cards:**

- **Industry:** Robotics & Physical AI — **Title:** Training Robots in Residential Indoor Spaces — **Text:** This 3DGS dataset provides detailed 3D scans of apartments and indoor scenes for robot training and physical AI applications. The dataset contains point clouds, 3D meshes, and spatial layouts captured from real-world environments. It helps robotic systems improve navigation, object interaction, and movement planning inside residential areas and complex indoor spaces.
- **Industry:** Computer Vision & Deep Learning — **Title:** Developing 3D Scene Understanding Models — **Text:** Researchers use this scan dataset to train deep learning models for object detection, scene reconstruction, and semantic understanding. The dataset includes detailed 3D models, point clouds, and reconstructed indoor scenes that support computer vision tasks. It enables accurate analysis of spatial relationships, furniture placement, and 3D object recognition in residential environments.
- **Industry:** AR/VR & Simulation — **Title:** Building Immersive Virtual Indoor Environments — **Text:** Developers apply this dataset to create immersive simulations and virtual indoor environments for AR and VR platforms. The 3D scans and textured meshes help reconstruct realistic residential spaces with accurate geometry. This supports simulation systems, digital twins, and interactive environments requiring detailed 3D layouts and real-world scanning data.
- **Industry:** Architecture & Digital Twin Systems — **Title:** Residential Space Reconstruction and Analysis — **Text:** Architectural teams can use this dataset to reconstruct indoor spaces and analyze residential layouts. The detailed 3D reconstructions support spatial analysis, digital twin generation, and visualization workflows. High-quality scanning data and structured 3D scenes allow professionals to study room geometry, optimize designs, and improve indoor modeling pipelines for residential projects.

### Fact

**FAQs Heading:** FAQs

**List of Questions:**

- **Question:** What formats are supported in this dataset? — **Answer:** The dataset supports multiple industry-standard formats, including PLY, LAS, OBJ, and GLTF.
- **Question:** How was the data collected? — **Answer:** Data was collected using the XGRIDS PortalCam scanning system in real residential environments. The capture process generates high-quality 3D scans, detailed meshes, and spatial reconstructions suitable for robotics and AI research.
- **Question:** What makes this 3D Residential Scans Dataset different from traditional point cloud datasets? — **Answer:** Unlike standard point cloud datasets, this collection combines 3D Gaussian Splatting (3DGS), high-quality meshes, multiple spatial representations, and motion sensor data captured with the XGRIDS PortalCam.
- **Question:** Does the dataset support robotics and physical AI training? — **Answer:** Yes, the dataset is optimized for robot training, robotics research, and physical AI development. The realistic 3D indoor scenes, point clouds, and accessibility layouts make it useful for navigation models, embodied agents, and autonomous robotic systems.
- **Question:** Does the dataset include real-world indoor environments? — **Answer:** Yes, the dataset contains scans of actual apartments with varying movement accessibility conditions. These realistic 3D indoor spaces are valuable for scene understanding, robot navigation, and simulation tasks.
- **Question:** Can I request a sample of the dataset before purchasing or downloading it? — **Answer:** Yes, sample files from the dataset can be requested for testing and evaluation.
- **Question:** How are Unidata datasets licensed? — **Answer:** Unidata datasets follow a dual-licensing model, where sample data is available for testing while full datasets are accessible exclusively through purchase. This allows organizations to evaluate the dataset before deployment in production or research workflows.
- **Question:** How are Unidata datasets stored? — **Answer:** Unidata stores datasets on secure AWS cloud infrastructure designed for scalability and reliability. Storage and management processes comply with ISO 27001 and ISO 27701 standards for information security and privacy management.
- **Question:** How long does it take to receive the dataset? — **Answer:** After submitting a request, the Unidata team reviews the project details and completes the required documentation process. Following agreement signing and payment, the dataset is typically delivered within 3–10 days.

[Full list of this site's AI-readable pages](https://unidata.pro/llms.txt)