It is amazing and fascinating to me every time I realize how many different fields of expertise need to converge to make 3D printing work. Design, material science and hardware engineering are just the main parts, there are so many more before, during and after phases: Authentise is about software driven 3D printing optimization and it is now using its expertise to give 3D printers better computer vision.
The field of computer vision itself is one that the 3D printing industry in general has only begun to really focus on, possibly because of the hardware and especially software requirements for it to provide real support rather than just increased complexity. It consists in enabling a computer to acquire, process and understand images in order to make decisions. In 3D printing this means understanding when a 3D print job is not being executed as planned and making the decision to abort in order avoid wasting energy and material. This video shows how it will work.
Authentise is a Singularity University supported start-up located at the NASA Ames Campus in Mountain View. So far the company has focused primarily on developing a cloud platform to stream designs to 3D printers in order for IP owners to preserve a certain degree of control on – and profit from – their 3D printable designs. Last year it launched the Computer Vision summer project, which involved the used of the OpenCV platform and has now come to fruition. It seeks to solve one of the biggest issues for many consumer, prosumer and professional adopters of 3D printing alike: unattended 3D print jobs gone bad.
All you have to do is mount the a camera on the 3D printers which uses Authentise cloud software to compare each new layer with the model. If the print fails the monitoring tools will send out an alert on your smartphone. You can then see the progress and decide whether to remotely abort the print job. All you need is a standard low cost webcam and you can even download a 3D printable mount from Authentise’s website. Can’t wait to try it out?. Sign up for the private beta here.
