computer vision

Math for Working with Rigid Bodies

Are you developing software and/or algorithms that need to represent objects situated and moving through 3D space? Are you modeling and/or measuring the attitude and position of devices, smart phones, IoT sensors, land, air, space, or sea vehicles, mobile robotic systems and/or moving mechanical components? Perhaps you’ll find this new book useful.

This started out as material to supplement a practical engineering discussion and evolved into this complete overview and practitioner’s guide now available in book form.

This technical brief summarizes an elegant algebraic description of rigid body position and attitude along with analytical derivatives associated with body motions. The equations presented here offer a fully general representation, with the minimum number of three free parameters while being entirely singularity-free. A particularly pragmatic benefit is that the math model and all of its parameters have direct associations with physical characteristics that can be observed and/or measured directly by physical instrumentation and sensor systems.

The creation of this technical note is an unabashed attempt to promote the use of geometric algebra (GA) for representing rigid body state and motion, and the use of bivectors for representing rotation in particular. Both are an effective methodology for representing and describing rigid body position and attitude for both static and kinematic situations.

The paperback book is available here or from most global book resellers including from: Lulu.com, Amazon.com, BarnesAndNoble.com, and elsewhere (e.g. search for >>>“Rigid Body Math” book<<<). The author would like to provide an e-book edition, but is in need of an effective tool for creating e-book formats that properly and fully capture the detail of LaTeX math expressions. If you’re desperate for an electronic version please email: info@stellacore.com with subject line “Rigid Body Math Model”.

Posted by webedit

Quick Camera Calibration

Photogrammetry and computer vision techniques are capable of providing astonishing levels of accuracy and precision when the equipment and processes are well calibrated. However, sometimes (fairly often actually), it’s only necessary to obtain reasonably good results, but to do so quickly. In these cases, there is no need, and no point, in spending extra time or effort (or money) on a high end camera calibration process.

This Stellacore technical note describes a simple camera calibration process, applied to a smart phone camera. The process is completed from scratch in less than 5 minutes yet produces results supporting 3D model reconstruction to better than centimeter level accuracy.

Posted by webedit