Services
Mirror reality. Master the future.
Scan to BIM
Digital Twin for Built Environment
Scan to BIM uses point cloud data captured via laser scanning, displayed clearly on Windows platforms. On this basis, paired with professional software, we build precise full-building models with accuracy ranging from LOD 100 to LOD 400, restoring real-world structures efficiently.
We leverage this workflow across urban planning, infrastructure construction, architectural interior & exterior design, and anomaly detection, providing reliable data support for smart urban projects and enabling high-precision digital delivery throughout the project lifecycle.
Road Design Workflow with Civil 3D
We build accurate road models efficiently using Autodesk Civil 3D. Start by importing field survey data to generate an existing terrain surface. Next, define horizontal alignment and vertical profile to establish the road baseline. Create standard road assemblies including lanes, shoulders and slopes, then generate a dynamic corridor model along the alignment.
The parametric model updates automatically with design changes, supporting volume calculation, visualization and construction drawing generation. It streamlines coordination across planning, design and construction, improving accuracy and delivery efficiency for infrastructure projects.
Bridge Design with Civil 3D
Designing bridges in Civil 3D starts with integrating topographic and survey data to build an accurate existing ground surface. We then define bridge alignments, approach roads, and grading layouts, while establishing parametric components such as abutments, piers, and superstructure assemblies.
By leveraging dynamic corridors and feature lines, the software automatically generates bridge models, cross-sections, and earthwork calculations. Changes to alignments or geometry update the entire model in real time, streamlining design validation, construction documentation, and project coordination. This workflow ensures precision, efficiency, and consistency for transportation and infrastructure projects.
Gaussian Splatting
3D Gaussian Splatting (3DGS) is a cutting‑edge 3D reconstruction and real‑time rendering tech unveiled at SIGGRAPH 2023. It uses millions of learnable 3D Gaussian ellipsoids to generate photorealistic 3D models from multi‑view images, with 100x faster training/rendering than NeRF, enabling real‑time display of fine structures, complex textures, and dynamic lighting. Widely used in digital twin, BIM modeling, cultural heritage digitization, VFX, and autonomous driving, it powers high‑precision, high‑efficiency 3D digitization.
Transform a single 360° video into a high-fidelity 3D model efficiently with Gaussian Splatting.
Start by capturing full-sphere footage to cover complete spatial context, then optimize frames, correct distortion, and estimate camera poses for consistent alignment.
The algorithm reconstructs the scene by optimizing millions of 3D Gaussian primitives—refining position, scale, rotation, and color—to faithfully reproduce geometry, lighting, and fine details without heavy mesh processing.
Mapping Factory
Point cloud data for plan drawing
In the application of 3D point cloud data, accurate mapping is a core implementation link! After completing professional classification based on 3D laser point cloud data, we use Mapping Factory mapping software to efficiently draw high-precision 2D and 3D floor plans, restoring the real scale and details of spatial scenes. The mapping results support delivery in multiple formats including DWG/DXF/SFC, adapt to the practical needs of various engineering design and spatial planning, and provide standardized for the subsequent project implementation.
Extraction of road markings
The key to road space digitalization lies in the accurate extraction and drawing of marking lines! Relying on 3D laser point cloud data and with the professional functions of Mapping Factory mapping software, we adopt the variable-width double-line technique to accurately outline the contour of road marking lines and perfectly restore the size, shape and layout characteristics of marking lines in different road sections. The extracted and drawn results can accurately meet the needs of road surveying and mapping and traffic engineering upgrading, and boost the digital modeling and refined management of road space.
Extraction of building lines
The digital modeling of building space starts with the accurate extraction of contour lines! Based on 3D laser point cloud data, we use the edge-by-edge collection tool of Mapping Factory mapping software to conduct refined collection and drawing for building edges, accurately restoring the contour shape, corner size and overall structural characteristics of buildings. The extracted and drawn building contour lines provide accurate basic graphic data for building surveying and mapping, BIM modeling, urban planning and other work, improving the efficiency and precision of building space digitalization.
Point Cloud
Case 1
What is Point Cloud Data
Point cloud data is a 3D spatial discrete point set obtained by 3D laser scanning, LiDAR, photogrammetry and other methods. It takes spatial coordinates (X, Y, Z) as the basic unit and can be attached with attributes such as color, intensity, time, echo times and so on.
Acquisition Methods of Point Cloud Data
In the field of 3D spatial data applications, point cloud data is the core foundation! Collected via 3D laser scanning, LiDAR, photogrammetry and other technologies, it takes X, Y, Z spatial coordinates as the basic unit, with additional multi-dimensional attributes such as color, intensity and echo count, forming a discrete point set in 3D space. We have diverse acquisition methods: ground, UAV, vehicle-mounted and aircraft-borne laser measurement, which can be flexibly selected for different scenarios, providing accurate and reliable data support for all types of 3D spatial analysis work.
Case 2
What is Point Cloud Filtering
Point cloud filtering is the process of removing unwanted parts from a point cloud, such as filtering out non-ground points and noise points.
Processing Flow
Preprocessing → Automatic Classification → Point Cloud Filtering → Inspection → Delivery.
Software Used
MicroStation, MapPowerView, TerraScan and TerraModerler.
Point cloud filtering is a pivotal step in point cloud processing! Simply put, it is a technical means to eliminate useless parts from laser-measured point clouds, including filtering non-ground points and removing noise that impairs data accuracy. We follow a standardized process: Preprocessing → Automatic Classification → Point Cloud Filtering → Inspection → Delivery, and use professional software such as MicroStation and TerraScan for operation. After filtering, the originally rough and messy ground point cloud data becomes smooth and standardized, greatly improving data usability and precision for subsequent work.
Case 3
Extraction of Buildings and Power Lines
Extract the point clouds of buildings and power lines into specified point classes respectively.
Accurate extraction of point cloud data is a key step in realizing spatial object analysis! In practical applications, we can filter and classify building and power line point clouds from the overall point set into designated categories through professional processing technologies. Targeted recognition and extraction based on their unique spatial features present clear 3D point cloud outlines of the two targets, providing precise data support for power engineering planning, building surveying and modeling, spatial planning and design, and boosting operational efficiency in related fields.
