Profile cover photo
Profile photo
CurvSurf
3-D object recognition, augmented reality, digital factory, 3-D data processing service, point cloud, orthogonal distance fitting, enabling technology
3-D object recognition, augmented reality, digital factory, 3-D data processing service, point cloud, orthogonal distance fitting, enabling technology
About
Communities and Collections
Posts

Post has attachment
FindSurface SDK - Revit Plugin Demo
Add a comment...

Post has attachment
FindSurface SDK - Simple GUI Demo
Add a comment...

Post has attachment
Sample Codes utilizing FindSurface 2018 are available at https://github.com/CurvSurf
https://developers.curvsurf.com/

- FindSurface-Sample-BasicSample: the simplest usage of FindSurface SDK
- FindSurface-Sample-SimpleGUI: an example of GUI application
- FindSurface-Sample-FindSurfaceRevitPlugin: Revit plugin source code.
Add a comment...

Post has attachment
Release of FindSurface 2018

3D Industrial Object Recognition SDK

On Jan 17, 2018, CurvSurf, inc. announced the release of FindSurface 2018 (www.curvsurf.com and https://developers.curvsurf.com/). FindSurface 2018, based on Advanced Applied Mathematics and Mechanics (AAMM), is an Intelligent 3D Industrial Object Recognition SDK, and it can directly utilize unfiltered Dimensional Point Cloud (or simply “Point Cloud”). It enables to bring Intelligent 3D object recognition to BIM, Reverse engineering, Intelligent Robotics, and immersive 3D applications into the everyday life.

CurvSurf provides three different types of licenses, e.g., FindSurface 2018 Trial, Standard, and Professional. Developers can sign up for free two-week trial of FindSurface 2018.

https://youtu.be/6tn4rMeIBaw

“The FindSurface SDK (i.e., FindSurface 2018) performs real-time, human-like recognition of 3D objects in point clouds. It works with point clouds generated by any device—from a terrestrial scanner to a consumer tablet with a 3D camera.” “The FindSurface SDK (i.e., FindSurface 2018) works a lot like the human brain. First, it reads the point cloud directly to pick out geometric “primitives” like boxes, planes, spheres, cylinders, cones, tori, and surfaces of revolution. After that, it calculates each object’s dimensional information—including position, orientation, size, radius, volume, and more.”

https://www.spar3d.com/sponsored/sponsored-software/findsurface-makes-real-time-automatic-3d-object-recognition-reality

FindSurface 2018 is to provide a continuous industrial-feature (topologies) recognition of the environment surrounding the users, in other words, “what is here”. This vision of the area around the user is what the user typically experiences. Most of 3D depth cameras, industrial 3D scanners, 3D vision systems, and 3D Lidar systems can be used with FindSurface 2018. Developers can easily incorporate FindSurface 2018 with their applications, e.g., automatic/semi-automatic feature recognition in reverse engineering, handling occlusions in AR/MR/VR, pallet & package measurements in logistics, localization or station registration, and many more in emergent industries. Initial release is supporting Windows, and other platforms will be supported in the future.

For more detail on FindSurface 2018, visit www.curvsurf.com and https://developers.curvsurf.com/.
CurvSurf
CurvSurf
curvsurf.com
Add a comment...

Post has attachment
3-D Augmented Reality - Apple ARKit

Virtual airplane rotates around the real gym ball with occlusion.

Sphere fitting to point cloud determines the position and size of real gym ball.
Add a comment...

Post has attachment
Add a comment...

Post has attachment
CurvSurf will release end Oct. 2017 the C# source code of the Revit Plugin for automatic feature extraction from point cloud
by using FindSurface SDK.

The C# source code is free for any usage including modification and distribution.
FindSurface SDK is free for one month testing.

Functions:
• Import point cloud
• Automatic feature extraction of plane, sphere, cylinder, cone, and torus
• Show/hide individual features, inliers, outliers
• Set control parameter values.

Usage:
1. Set program control parameter values:
- Point measurement accuracy
- Touch radius: Size of point cloud patch
- Mean distance: Point cloud density
- Threshold of cone’s vertex angle
- Region growing: Radial expansion, lateral extension.
2. Select feature shape to be extracted: Plane, sphere, cylinder, cone, torus, or any
3. Click a point among the point cloud
4. Repeat again from 1, 2, or 3.
PhotoPhotoPhotoPhoto
17. 9. 27.
4 Photos - View album
Add a comment...

Post has attachment
Unspoken Reality to Audience

1. AR technology is mainly based on plane detection as of table top, floor, wall, street, etc.
2. AR demo videos are recorded under well controlled choregraphy because of occlusion problem.

Without a solution to, the hyped-up expectation of audience will get be disappointed.

Possible solutions could be developed based on point cloud without meshing.
Add a comment...

Post has attachment
Add a comment...

Post has attachment
We have experimented on ARKit with an iPad Pro.
Even though the point cloud generated is highly noisy, CurvSurf FindSurface extracted successfully a sphere for gym ball and a plane for floor. The extracted sphere and plane are in contact as real.
Add a comment...
Wait while more posts are being loaded