Monocular Multi-Directional Camera Room Mapping Without Depth Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robotic devices face challenges in effectively mapping and navigating enclosed spaces due to limited perception and computational resources, as well as the cost and complexity of specialized sensors required for accurate three-dimensional mapping.
Innovation Solution
A method using a monocular multi-directional camera to estimate dimensions of enclosed spaces by obtaining image data, determining pose data, evaluating volumetric functions for depth values, and fitting three-dimensional volumes to these values to determine polygonal cross-section dimensions, which can be used to map and navigate spaces efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If specialized sensor devices such as LADAR, structured light sensors, or time-of-flight depth cameras are used, then three-dimensional mapping accuracy is improved, but device cost and complexity increase
Solution Approach 1:
The patent uses a monocular multi-directional camera to capture images from multiple angular positions, creating multiple two-dimensional image copies that are processed to reconstruct three-dimensional space. This approach replaces expensive specialized sensors with a simpler camera system that captures multiple views, achieving depth estimation through computational processing of these image copies rather than direct depth measurement.
Solution Approach 2:
The patent replaces mechanical/optical depth sensing systems (LADAR, time-of-flight cameras) with a computational vision system. Instead of using specialized sensors that directly measure depth through physical principles, the system uses a standard camera combined with multi-directional imaging and computational algorithms to estimate three-dimensional dimensions, substituting physical measurement mechanisms with information processing.
2Measurement precision
If research laboratory computational resources are used, then SLAM techniques achieve high accuracy, but embedded computing devices lack sufficient processing power
Solution Approach 1:
The patent extracts and isolates the essential computational steps needed for dimension estimation from complex full SLAM systems. Instead of implementing complete simultaneous localization and mapping algorithms that require substantial processing power, the system extracts the specific functionality needed for estimating enclosed space dimensions, reducing computational requirements while maintaining accuracy for the specific task.
Solution Approach 2:
The patent performs partial SLAM functionality by focusing only on the aspects necessary for dimension estimation rather than complete environmental mapping and localization. The system captures images from multiple angular positions and processes them to estimate space dimensions, performing just enough computational work to achieve the specific goal without the overhead of full SLAM implementation.
3Device complexity
If a monocular multi-directional camera is used, then device cost is reduced, but depth estimation capability is limited
Solution Approach 1:
The patent compensates for the limitations of a monocular camera by introducing the dimension of multiple angular positions. Instead of relying on a single viewpoint, the system captures images from multiple directions around the enclosed space, using the angular variation to infer depth information that a single-camera viewpoint cannot provide, effectively adding a dimensional aspect to the imaging system.
Solution Approach 2:
The patent introduces computational processing as an intermediary between the monocular camera and depth estimation. Rather than relying on the camera alone to directly measure depth, the system uses image processing algorithms as a mediator to analyze multiple images from different angular positions and compute depth values, bridging the gap between simple imaging and accurate depth measurement.
Data Source
AI summary
Certain examples described herein relate to estimating dimensions of an enclosed space such as a room using a monocular multi-directional camera device. In examples, a movement of the camera device around a point in a plane of movement is performed, such as by a robotic device. Using the monocular multi-directional camera device, a sequence of images are obtained at a plurality of different angular positions during the movement. Pose data is determined from the sequence of images. The pose data is determined using a set of features detected within the sequence of images. Depth values are then estimated by evaluating a volumetric function of the sequence of images and the pose data. A three dimensional volume is defined around a reference position of the camera device, wherein the three-dimensional volume has a two-dimensional polygonal cross-section within the plane of movement. The three dimensional volume is then fitted to the depth values to determine dimensions for the polygonal cross-section. These dimensions then provide an estimate for the shape of the enclosed space.