What is Photogrammetry and How Can You Use it to Create 3D Models?
Photogrammetry is the process of taking multiple pictures of an object and stitching them together to create a digital 3D model. There are various software programs available for photogrammetry, which have continued to improve with the advent of artificial intelligence and machine learning. Some of the best photogrammetry software out there are:
No matter which photogrammetry software program you choose, they all work on the same basic principles.
Getting Started with Photogrammetry
First, it is important to have a high-quality camera when taking pictures of an object for photogrammetry. The quality of the resulting 3D model is dependent on the resolution of the photographs. Today most cell phones can take 4K resolution photographs so it is easy to get started without purchasing expensive equipment.
Capture Your Images
It is also essential to keep the camera steady at all times. You can opt to set the camera on a tripod and use a turn-table for the object so you can take photographs from every angle easily. Professional photogrammetry studios usually have robotic arms that move the camera around to every angle required. A steady position of the camera at multiple angles is important because photogrammetry software uses triangulation to determine the shape of the object. The position and angle of the camera in relation to the object allows the photogrammetry software to calculate the exact point in space of the object.
Generating the Required Files for 3D Photogrammetry
Doing this repeatedly around the product enables the software to create a point cloud that represents the shape of the object. The point cloud is then converted into polygons that represent an accurate 3D geometry of the object. Then the model is mapped with the materials and textures generated from the images. This typically results in an OBJ file with JPGs or PNGs and an MTL.
- The OBJ is the 3D geometry of the object.
- The JPGs or PNGs are the materials and textures
- And the MTL file dictates where the materials and textures map to the geometry.
So, What types of objects can photogrammetry capture?
Photogrammetry can be used to capture a wide variety of objects as small as a pen to full buildings and more.
- Architecture & Land Surveying: Satellites and drones are used to take images from multiple angles of buildings and large areas of land. The real estate industry uses this to provide 3D models of homes or buildings to potential buyers as well as survey land for development. The military also uses aerial photogrammetry to map out large areas of the earth. There really is no limit to the size of a surface you can map with photogrammetry.
- Products: 3D models of products can be used across eCommerce and social media, which has proven to double conversion rate, increase cart size and reduce returns. In the past, 3D models were time consuming and expensive to create because they required a lot of manual work by 3D artists to sketch the object by hand. Technologies like photogrammetry and 3D scanning help automate this process, making it more feasible for companies of all sizes to leverage 3D across sales and marketing channels to engage more customers.
What is the difference between 3D Scanning and Photogrammetry?
The term 3D scanning is often used interchangeably with photogrammetry, but they are really two different methods of capturing 3D models. 3D scanning is the process of using lasers to map an object’s shape and create a 3D digital model. There are a variety of lasers that can be used for 3D scanning, but they all also use the principles of triangulation to map an object.
In general, 3D scanners shoot laser light at an object and have sensors to measure the time it takes for the light to bounce back (aka time of flight) to calculate the distance from the camera to the object which is then used to map the shape. This technology can achieve a very accurate shape and is ideal for reverse engineering parts and creating 3D models of very detailed products. It does not, however, capture the material or texture of an object like photogrammetry does.
What is the best 3D model capture solution?
The best 3D model solutions use both photogrammetry and 3D scanning to achieve the highest quality 3D models. Below are an examples of high quality 3D shoes that were created using a combination of 3D scanning and photogrammetry:
How can you get started with 3D Photogrammetry today?
VNTANA works with 3D scanning experts around the world to create these high-quality 3D products in convenient locations for all of our clients. If you are a brand looking to digitize your products for eCommerce and social media, professional 3D scanning and photogrammetry studios are the way to go for now. However, the technology is continuing to get better with the cost of cameras and lasers dropping and photogrammetry software improving with artificial intelligence and machine learning to better predict the shape and look of an object.
We’re extremely excited about the latest Apple announcement at WWDC this year on Object Capture. Object Capture will enable anyone with an iPhone or iPad to create a realistic 3D model with photogrammetry using Apple’s RealityKit 2 SDK. This allows developers to utilize Apple’s photogrammetry software to create 3D models in USDZ format which are ready for iOS based augmented reality. You must download macOS Monterey beta and Xcode 13 beta. You will need an iPhone 12 Pro, Pro Max, or 2020 iPad Pro as these all have LiDAR scanners built in which improve the capture quality. You can find more details to get started here:
https://developer.apple.com/augmented-reality/object-capture/
If you’re not a developer, but want to start capturing 3D models with your iPhone you are in luck as there are several apps that have been released on the app sotre that take advantage of this technology. Rumor has it that there should be another big announcement by Apple this Fall that might make this process even easier for amateurs with an iPhone and we can’t wait!
Whether you are building your own photogrammetry tool with RealityKit 2 or using an app based on the SDK, you will need additional tools to use the 3D file across web, mobile, social media and game-engines. Once you have a USDZ scan of your product and you want to use the 3D model on your website, you will need a GLB or glTF model as that is the standard format for 3D on the web and Android AR. You will also need to optimize the 3D model to ensure the file size is small enough and the polygon count is low enough to load fast across web and mobile. VNTANA automatically optimizes and converts 3D files so you don’t have to worry about this manual work. We also provide a 3D web viewer with built in AR that you can easily embed on an eCommerce website using an iFrame or our API. And if you don’t want to bother with any of it, our 3D experts at VNTANA can handle everything for you. Just ship us your product and we can have your 3D products live in less than a week.
