No-Code & Low-Code Computer Vision – How It Works and Why You Need It

TLDR: Computer Vision (Visual-AI) is an incredibly powerful tool that can power an almost limitless number of applications. It has traditionally required integration using an API, which has limited its use in many areas where companies either did not have a platform to integrate it with, had no need to integrate it, or did not have the coders to do it. But now, a new approach called ‘No-Code Computer Vision’ provides the benefits of Visual-AI without the need for even one line of code to be written.

No-Code & Low-Code Computer Vision – How It Works and Why You Might Need It

APIs have been around almost as long as computing itself, albeit in much simpler forms. The modern era of APIs started in the early 2000s thanks to three companies, Salesforce, eBay, and Amazon who each launched their API suites to open up their platform to third-party developers. 

Since then the popularity of APIs has grown to the point that no self-respecting platform would ever launch without a corresponding API and today APIs power almost every aspect of what we do, allowing us to connect with and control cloud applications and data, and interface with hardware.

But, as powerful and enabling as they are, APIs are not always the best way for some companies to get what they need. In this article we look at two alternative approaches and why they might just be what you need.

So, You know how Computer Vision APIs work?

The fundamental basis of APIs is to allow systems to send a request to another system and get a response. That response will typically be in JSON or XML format (although other formats are also emerging for specific applications).

In our world, our customers use the VISUA API to send us visual media (images and videos) in order to have our computer vision engine process that media using the various technologies in our suite (from logo and text detection to object & scene detection and visual search). The API allows the user to define many aspects of how they want the media processed, which is then reflected in the data output they get back.

Typical customers for our API include cybersecurity, social media monitoring, counterfeit detection, and product authentication companies. They all have one thing in common…they either have a platform or an internal software system that needs to embed the data we extract for further analysis by, or presentation on, their software.

OK, great, so APIs are the answer to everything right?… Wrong!    

Why You Might Not Want To Use An API 

There are some applications for third-party systems where the results of the processing are highly desirable, but the raw data itself is not. Implementing APIs is typically very easy, but figuring out what to do with the data when you get it back is less so. Not only do you have to filter and format it, but you have to define some sort of interface or dashboard, or integrate it into your existing UI to display the outcome.

So, what if you don’t want to do any of that, but you do want visual content processed? That’s where our ‘Low-Code’ and ‘No-Code’ options come in.

These options are perfect for companies that:

• Don’t want to integrate the output data into their platform
• Don’t need to combine the data we extract with their own data
• Where the ultimate result is ‘binary’, i.e. yes, there is an element/issue or no there is not.

To understand how they work we need to go back to the concept of APIs having two elements to them… being the sending and receiving of instruction/data. The Low/No-Code options eliminate the receiving of data from our API to them. Instead, clients make use of a dashboard provided by us where we display the output of the visual processing according to filters and rules that the client stipulates.

The perfect example of this is platforms that need to moderate their visual content. From social media platforms to print-on-demand companies that are making a decision about the validity/appropriateness of visual content.

But, that still leaves the sending of data and instructions to our engine for processing and this is where the two options cater to different needs:

1. Low-Code Computer Vision

Low code allows you to use the API to automate the sending of data to our engine for processing and while doing so, you can include specific corresponding data that links the image or video to a specific post or order on their system. Once processed, the image/video is displayed in our dashboard with the identified elements and a determination for action based on your rules. The extra data provided by you is then displayed with each visual element, allowing you to reference it back to the original post/order.

2. No-Code Computer Vision

For those who simply don’t want to touch our API (perhaps because they don’t have coders on staff), our no-code option is the answer.

Just load files that you want to be processed into a specific location (FTP folder, Cloud Drive, Bucket) and we will collect them on a regular basis that can be as often as every minute to every day. On your side, you can again add metadata to the files themselves, which when processing, we extract for adding to the dashboard during output.

Although this approach is ideal for lower volume requirements, the combining of metadata into the files and storing to the pick-up location can be automated on your side where volumes grow.

Want To Talk?

These two options open up the possibilities of computer vision for companies with a broad range of needs and capabilities and not just those with a team of developers and the capacity on their roadmap to make it all happen.

If you’d like to learn more about our various API, Low-Code or No-Code options for accessing our Visual-AI (computer vision) technology just fill in the form below.