We're Objective. We're an AI-native search platform for developers. If you're building a search app, grab an account today and check us out!
Imagine searching for this photo in a collection of digital assets:
Traditionally, someone would need to tag this photo with keywords: "child", "cat", "skateboard", and whatever else comes to mind, then hope these keywords match whatever search terms are used. This is tedious and time-consuming. With Objective's Image Search API, building search into your web or mobile app that understands language and sees inside images is easy.
If you're using some cloud platform, most of them either:
- Put the burden on content creators. On some platforms, the person creating or uploading the content is responsible for enriching it with appropriate text/tags so that the keyword search system can find them. They're already busy and labelling photos isn't their speciality, so this extra work on their plate can lead to bad or incomplete tagging.
- Hire a professional tagging team. Since user-generated tags cannot be trusted, many platforms employ professional services that have trained staff to tag at higher accuracy levels. This is expensive and does not scale well since the workforce churns and training takes time.
- Use automatic tagging (AI) services. A more scalable approach is to train an AI system that can quickly and cheaply tag large amounts of data. This scales better and can be successful to a degree, but it is also restrictive since one cannot possibly guess all the different ways users will search for assets.
In the photo above, there are
- Colors (what if you specifically want photos with "orange cats"?)
- Textures
- Scene descriptions
- Different possible terms for the same thing (what if someone searches for "kid" instead of "child"?)
- And even sentiment and action ("happy", "playing")
If a picture is worth a thousand words, then coming up with a list of keywords to describe all the possible information in this photo is almost impossible. The result? Wasted media assets that don't get discovered, and duplicate work from recreating what already exists.
How about we skip the tagging and search image content directly?
At Objective, we've built a media understanding platform that helps you search your images the way you think and speak. The platform deeply understands the content of your images, and lets you search them using natural language.
And the best part? You can say goodbye to tagging. Just upload your digital assets, and you're done. Our engine will take care of ingesting and indexing your images, and offer a simple search API to help you find the perfect image every time.
Here's an example of how our AI platform performs against conventional tag-based search on a library of stock images. None of the images in the results we surfaced required tagging with words like "red", "brick" or "house".
It just works.
But don't take our word for it. Try it for yourself here.
And see how to build it in just a few minutes with Python.
A year ago building something like this would have required a PhD and writing a ton of machine learning code. But now you can leverage AI-powered search for your digital assets in a few clicks. If you have images or other content you’d like to make searchable we’d love to help you, please get in touch!
