Now, one Seattle startup is doing something similar — with a little less clothing.
Algorithmia, a Seattle-based startup that builds a marketplace for algorithms, just debuted isitnude.com (safe for work), which can scan an image and detect whether or not the people in the photo are wearing clothes.
The company describes how the software works in this blog post. It started with existing programs that look for skin patches in a given photo. However, this method may mistake a beige dress or hand as “nude,” so Algorithmia took it a few steps further.
Here’s how the company explains its process:
Our contribution to this problem is to detect other features in the image and using these to make the previous method more fine-grained. First, we get a better estimate of skin tone using a combination of what we learned from Human Computer Interaction Using Hand Gestures and what we can get from the image using a combination of OpenCV’s nose detection algorithm and face detection algorithm, both of which are available on Algorithmia. Specifically, we find bounding polygons for the face and for the nose, if we can get them, then get our skin tone ranges from a patch inside the former and outside the latter. This helps limit false positives. Face detection also allows us to detect the presence of multiple people in an image, thus customizing skin tones for each individual and re-adjusting thresholds so that the presence of multiple people in the image does not cause false positives.
Algorithmia built the isitnude.com website to demonstrate what’s possible on its platform, which helps developers use algorithms in their own apps with fewer than 10 lines of code by using Algorithmia’s API. The company wants to make it easy for computer researchers to take algorithms that would ordinarily stay locked up in academic papers and monetize them — making them available to businesses who want to take advantage of new technology.
Algorithmia launched earlier this year and raised a $2.4 million seed round from Madrona Venture Group, Rakuten Ventures, Deep Fork Capital and a group of angel investors including former University of Washington computer science professor and Farecast founder Oren Etzioni and former Microsoft general manager Charles Fitzgerald.
Speaking of photo detection, USA Today reported on Tuesday how Facebook is experimenting with its own software that can detect faces in photos that are obstructed or hidden.