Matterport, Inc.

Google Inc.

Mykhaylo Kurinnyy

Matthew Tschudy Bell

Gunnar Hovden

David Alan Gausebeck

Daniel Ford

Jonathan James Taylor

Edward T Chang

Zhaowei Wang

Haakon Erichsen

Techniques are provided for increasing the accuracy of automated classifications produced by a machine learning engine. Specifically, the classification produced by a machine learning engine for one photo-realistic image is adjusted based on the classifications produced by the machine learning engine for other photo-realistic images that correspond to the same portion of a 3D model that has been generated based on the photo-realistic images. Techniques are also provided for using the classifications of the photo-realistic images that were used to create a 3D model to automatically classify portions of the 3D model. The classifications assigned to the various portions of the 3D model in this manner may also be used as a factor for automatically segmenting the 3D model.

Automated classification based on photo-realistic image/model mappings

This document describes techniques and systems that enable autocalibration for multiple cameras using near-infrared (NIR) illuminators. The techniques and systems include a user device that uses NIR images, including dot images and flood images, captured by the multiple cameras. The user device implements an autocalibration module that normalizes the contrast of each image. Then, the autocalibration module detects dot features in the dot images and detects flood features in the flood images. The autocalibration module uses the flood features to disambiguate the dot features in the dot images. Then, the autocalibration module uses the disambiguated dot features to determine new calibration parameters for recalibration of the multiple cameras. In some aspects, the autocalibration module uses the dot images, rather than the flood images, to detect the flood features for disambiguating the dot features.

Autocalibration for multiple cameras using near-infrared illuminators

Systems and methods for generating three-dimensional models having regions of various resolutions are provided. In particular, imagery data can be captured and utilized to generate three-dimensional models. Regions of texture can be mapped to regions of a three-dimensional model when rendered. Resolutions of texture can be selectively altered and regions of texture can be selectively segmented to reduce texture memory cost. Texture can be algorithmically generated based on alternative texturing techniques. Models can be rendered having regions at various resolutions.

Optimal texture memory allocation

Systems and methods are provided for automatically separating and reconstructing individual stories of a three-dimensional model of a multi-story structure based on captured image data of the multi-story structure. In an aspect, a system is provided that includes an analysis component configured to analyze a three-dimensional model of structure comprising a plurality of stories generated based on captured three-dimensional image data of the structure and identify respective stories of the plurality of stories to which features of a three-dimensional model are associated. The system further includes a separation component configured to separate the respective stories from one another based on the features respectively associated therewith, and an interface component configured to generate a graphical user interface that facilitates viewing the respective stories as separated from one another.

Classifying, separating and displaying individual stories of a three-dimensional model of a multi-story structure based on captured image data of the multi-story structure

Growing community of inventors

Milpitas, CA, United States of America

Mykhaylo Kurinnyy