MVTec Software GmbH

3D Registration & Perspective Matching

Featured in HALCON

3D Registration

3D sensor determining the shape of an object
Using 3D registration, the three-dimensional object shape is determined from multiple views with a 3D sensor.
3D shape of an object identified with 3D registration
The three-dimensional object shape is determined with 3D registration.

3D sensors directly capture the 3D coordinates of the visible points on the object surface. Using multiple 3D images of the object acquired from different points of view, the pairwise overlap of the 3D data is computed and a globally optimized shape is derived.

This technology is particularly suitable for 3D reconstruction, i.e. determining the 3D shape of arbitrary objects. The 3D object shape can be used in 3D matching, for quality inspection of 3D objects, or for the position recognition of 3D objects.

Our software packages support the direct derivation of the 3D object shape from 3D sensor data using 3D registration. Furthermore, our software offers advanced 3D shape processing techniques like smoothing, subsampling, and triangulation to prepare the shape model for 3D matching and object comparison.

Surface Fusion For Multiple 3D Point Clouds

Surface fusion for multiple 3D point clouds
Surface fusion

HALCON offers a method that fuses multiple 3D point clouds into one watertight surface. This method is able to combine data from various 3D sensors, even from different types like a stereo camera, a time of flight camera, and fringe projection. This technology is especially useful for reverse engineering.

Perspective Matching

A camera determining the 3D pose via a planar part of the object
Only a planar part of the object is necessary to determine the 3D pose of the object by exploiting perspective matching techniques.
Determining the 3D pose of five objects by finding the trained deformable model
Calculating the 3D pose of objects by finding the trained deformable model.

Instead of using the full 3D shape of an object, for many applications it is possible to restrict the model area to a planar part of the object. For arbitrarily shaped object parts, perspective matching allows to determine the 3D pose with only one camera. The model generation is done by training a sample image of the object typically inside a specified ROI.

Applications for this technology can be found within 3D alignment, e.g., applications for which the 3D pose (position and orientation) of an object must be found. Examples for such would be automotive and robotics applications, pick-and-place applications, and bin picking. A further possibility is the measurement of geometric features on complex 3D objects after 3D alignment.

Our software's perspective matching helps finding objects easily with only one camera. For perspective matching, our software packages provide two different methods suitable for two different classes of objects. Depending on the object's shape and appearance, our software offers both deformable matching, which is based on the shape-based matching technology (object edges), and descriptor-based matching, which uses so-called interest points.

Read Success Stories about 3D Vision


Vision system sorts tomato seedlings - University of Wageningen

More than 150 million tons of tomatoes are produced every year around the world. With seedlings costing €0.25, plant growers such as Westland...

Automotive & Robotics

Electrode sheet inspection machine - Ayaha Engineering

For web inspection, a high resolution line scan camera is basically used. To stably process large images in high speed, a fast hardware processing...

Automotive & Robotics

Vision-guided robotic 'bartender' serves drinks - University of Brescia

Researchers led by professor Giovanna Sansoni at the Laboratory of Optoelectronics in the Department of Engineering at the University of Brescia,...


Machine Vision Inspects Biscuits – Machine Vision Technology Ltd.

As imaging software becomes more advanced and the latest 3D imaging tools open the doors to techniques that where not possible before, one of Europe's...

Automotive & Robotics

Vision-guided Robot Heads for Space – NASA

To support the work of astronauts on the International Space Station (ISS), NASA (Washington, DC), General Motors (Detroit, MI) and Oceaneering Space...

Automotive & Robotics

Novel gripper uses machine vision to automate packing - Helmut-Schmidt-University Hamburg

Combining a sophisticated gripping mechanism with machine vision allows packing systems to handle disparate objects


Structured-light-based vision system inspects percussion caps – Tekniker

The Centro Tecnológico Tekniker has developed a very fast system to inspect percussion caps.

Percussion caps are a critical component used in...


Vision System Simplifies Robotic Fruit Picking – ACRO Institute

Robotic apple picker relies on a camera inside the gripper and off-the-shelf software


Automotive & Robotics

3D Calibration for Inspection

Stereo camera system eases highly precise measurement