For scuba diving enthusiasts, mask comfort and fit are paramount for an enjoyable and safe underwater experience. Mask leaks and discomfort can quickly ruin a dive, turning a serene exploration into a frustrating ordeal. Driven by personal experience and a passion for innovation, I embarked on a project to create custom-fitted SCUBA masks using the power of 3D printing. As a former Navy submariner and diver, I understand firsthand the critical need for reliable and comfortable gear, especially during extended dives. My background in product design and development, combined with access to cutting-edge technology like structured-light 3D scanners and 3D printers, has fueled this endeavor. The goal is to develop a SCUBA mask that is not only perfectly contoured to the diver’s face, eliminating leaks, but also provides unparalleled comfort for long-duration dives. This article will document the journey of creating these personalized 3d Printed Masks, starting with the crucial first step: capturing accurate facial data for a flawless custom fit.
One of the initial hurdles in producing custom SCUBA masks is obtaining precise facial data from individuals to generate an accurate and usable 3D model. My first attempt involved utilizing “FaceGen,” a software designed to create 3D facial models from photographs, typically requiring frontal and profile views.
Example of FaceGen software interface for 3D face modeling from photos
While FaceGen produced visually appealing models that resembled the subject, a closer examination of the 3D mesh revealed a generalized shape rather than a truly accurate representation of the individual’s facial contours. FaceGen’s primary purpose is to generate realistic-looking faces for applications like gaming and animation, relying on statistical averages of human facial features. Despite its initial promise, FaceGen proved unsuitable for the precise custom-fit requirements of a SCUBA mask. Developing a proprietary software solution for 2D-to-3D conversion was considered, but deemed too extensive for this project’s scope. The search then shifted towards existing software capable of converting 2D images into accurate 3D models, leading to the discovery of Autodesk’s 123D Catch (now known as ReCap Photo).
Autodesk 123D Catch, a free application, offered a promising solution for creating 3D models from a series of photographs taken around a stationary object. The quality of the 3D model is directly proportional to the quality and number of photographs. For this test, I used a standard Canon point-and-shoot camera in good lighting conditions. My wife rotated around me, capturing a series of high-resolution images while I remained perfectly still, maintaining a roughly equidistant position. A ruler with one-inch circles was included in the scene for later model scaling. These photographs were then uploaded to 123D Catch, where Autodesk’s servers processed the images to generate a 3D model. Below is a screenshot of the raw 3D model produced by 123D Catch, ready for export in standard 3D mesh formats.
Raw 3D model output from Autodesk 123D Catch software
The initial 3D model generated by 123D Catch was impressive, allowing for rotation and zooming within the software. The next step was to find a program capable of exporting this data into a usable format for further manipulation and 3D printing. MeshLab, a free and open-source software for viewing and manipulating 3D meshes, proved to be the ideal tool.
3D models are represented by meshes, intricate networks of hundreds of thousands, or even millions, of interconnected triangles. MeshLab allowed for the visualization and manipulation of this mesh data. The image above showcases the facial mesh and a rendered 3D model after some initial trimming. The level of detail captured was remarkable, clearly outlining facial features such as the under-nose contours, brow bone, and cheekbones. While minor imperfections like a slight bump on the nose were present, the accuracy was deemed sufficient for creating a custom-fit SCUBA mask. To validate the dimensional accuracy of the 3D model, physical measurements of prominent facial features, such as the distance between the cupid’s bow and the distance between pupils, were taken for comparison.
However, scaling and warping the model to achieve precise dimensional accuracy required more processing power and specialized software than readily available. To overcome this, I utilized Upwork.com, a freelance platform connecting businesses with skilled professionals.
Upwork offers a vast pool of talent, including 3D modelers, website developers, and designers. For a modest fee of $25, a freelance 3D modeler efficiently trimmed, scaled, and validated the dimensional accuracy of the model within a day. They also converted the mesh file into the .STL format, essential for 3D printing. This service was significantly more cost-effective compared to a $450 quote from a traditional 3D modeling company for the same task. The resulting completed 3D model, ready for 3D printing, is shown above, loaded in Netfabb, a free software for validating 3D print models.
In summary, the process of generating an accurate 3D facial model from photographs was successfully achieved using a combination of free software and freelance services. This approach enables remote data acquisition from customers, a crucial aspect for creating custom SCUBA masks on demand. The lean start-up methodology, prioritizing cost-effectiveness, has been central to this project, leveraging tools like 123D Catch, Meshlab, Upwork, and Netfabb.
The next critical question was the cost of 3D printing a life-size model of the head. Traditional 3D printing services quoted upwards of $1000, a prohibitive cost for prototyping. This highlighted the transformative impact of the “maker” revolution and online communities that empower individual projects. Platforms like 3DHubs.com connect individuals with 3D printers to those needing printing services. For just $40, a 3D printing enthusiast in Kansas printed the facial model in ABS plastic, followed by acetone vapor polishing for a smooth finish.
Most consumer-grade 3D printers utilize Fused Deposition Modeling (FDM). This process involves extruding heated plastic filament layer by layer onto a build platform. FDM printing can result in visible layer lines due to printer resolution limitations. While these lines can be smoothed mechanically, acetone vapor polishing offers an alternative for ABS prints. Acetone vapor polishing involves exposing the 3D print to acetone vapor, which melts the surface layers, creating a smooth finish while preserving intricate details. The image below showcases the acetone vapor-polished 3D printed face model.
While printing every customer’s head isn’t feasible, this initial print served to validate the accuracy and dimensional fidelity of the 3D data. Measurements from this print will be used to refine the mask design and create clay models for fit testing on an actual face. A sneak peek of the custom SCUBA mask model is provided below.
This initial technical exploration demonstrates the potential of leveraging open-source digital technology and community-driven platforms like Upwork and 3DHubs to create personalized 3D printed SCUBA masks. The journey continues with the design and 3D printing of the mask itself, building upon the accurate facial model created. Stay tuned for future updates on the mask design process and the realization of comfortable, leak-free, custom 3D printed masks. For more details and ongoing progress, visit www.reeftoridge.com. Questions and comments are welcome in the comments section or via direct message.
