One of the best strategies for innovation is to start by giving yourself some boundaries. We like to make a game of it. As the saying goes, you don’t need to try to boil the ocean. In this series, we begin by restricting our game to a thought starter that has some known benefits and limitations to get us going.
What is Computer Vision?
Everyone has heard of “artificial intelligence” or even perhaps “machine learning” at this point. At a high level, both encompass of field of study and thought devoted to giving a computer program some input and having the computer “understand” what it is to the point that it can return information.
I like to think of it as gaining a super-power.
Computer vision is no different. So let’s start our innovation game by looking at some of the ways computer vision can be applied. To paint a picture of the scene in your mind, assume that the computer we are talking about is a phone that has its camera activated. Some of these are general skills and others are very specific.
- Identifying letters or words in the camera frame
- Identifying that a person’s face is in the camera frame
- Identifying the SPECIFIC person’s face in the camera frame
- Recognizing any of the widely used barcode types - QR code, UPC code, ISBN, etc. Even Braille.
- Recognizing movement
- Tracking movement
- Making a best guess at an object. (A computer can be trained to tell the difference between a dog and a house in the camera frame)
- Finding lines in an image
- Gesture recognition
- Augmented reality. (Placing graphics on top of the live camera feed)
Do any of those spark an idea about something you could apply to your business?
Let’s look at a simple one - recognizing a shape or a line in the camera frame. Autonomous driving is made possible in part by the car’s ability to recognize the lines on the road. By interpreting where the car is relative to the lines, the system can send commands to the car’s wheel to stay inside the lines.
Another example is that line you see in football games showing where the first down is. The computer can easily pick up on where the yard markers are on the sidelines, add ten yards and draw a yellow line in perspective. In recent years, the broadcasts have grown to add team logos, stats and other data on top of the feed to add context.
What if you gave that super power to your company? How would you use it?
Scanning a barcode is an interesting one. This technology has been around forever, but with the ability to combine it with augmented reality, it can provide incredibly interesting solutions. Imagine a person working in a warehouse walking up to a bin with a barcode. The person holds the phone up to the code and the system can ask questions. Does this need to be restocked? How many are in stock now? Are you taking part with you? How many? The system can also lay other data on top of the screen based on what the person is looking at. Is it helpful for the worker to see what other parts go with this one? What assemblies require it?
To come up with something valuable, start small. Narrow your focus. Solutions like these can build and evolve over time. Start with something that can make a difference today and then iterate on it.
