As I told in the article The blocker of the restless developer, I have been worked on image.ai, an artificial intelligence app which discovers all the elements of an image.
At least for now, I’m not going to continue moving forward the project because I do not have the strength to start a big project on my own. But I want to share current status.
Leaving aside, the most technical part, in addition to the basic functions such as the home, authentication, user account, … the goal of the app is to receive images and analyze them.
Once we have the images, we can do some analysis: find faces, labels and logos.
I have worked with AWS Rekognition and Google Vision (also, I have created some small libraries to work with: node-rekognition and node-google-vision and the code it is ready to receive the rest of analysis of this platforms, both frontend and backend.
After the analysis, you can filter the images, with progressive filters on the frontend. For example, if you select some feature (iPhone in this case), the rest of the images without this feature will disappear. And you can add all the filters that you want.
Now, the app receives images from a file input, but it could be a Twitter account or hashtag, websites, … or videos. For example, you could find in which moments of a video a specific face appears, see what brands use the people that appear on the images of a Twitter hashtag or spy the media content of your company competitors.