Technology

It might seem like magic.
But really, it's just science and math.

Clarifai was one of the first companies to offer Artificial Intelligence solutions, transforming how businesses and developers build apps and create user experiences. Learn how we solve real-world problems.

What is computer vision?

Computer Vision is the ability to automatically understand any image or video based on visual elements and patterns. Clarifai applies machine learning to image & video recognition, helping customers understand and manage their media.

How does Clarifai's computer vision technology work?

 

Clarifai uses deep convolutional neural networks (CNNs), a subfield of machine learning, to understand objects in images and videos. We developed a proprietary, state-of-the-art neural network architecture, and trained the network over billions of training samples to offer world-leading models for our customers.

 

Our machine learning models are powered by Clarifai Object Recognition Engine (CORE), which adds millions of images to its repository on a daily basis and retrains our models to provide the most accurate results.

Benefits of the Deep Learning Approach

Scalable

Deep neural networks scale to billions of parameters giving them capacity to learn highly complex concepts and thousands of categories. With modern hardware and abundance of data we are able to train larger and more powerful networks.

Fast

A trained model stores its knowledge compactly in learned parameters, making it easy to deploy in any environment. There is no need to store any additional data to make predictions for new inputs. This means that we can easily use them on embedded devices and provide responses in milliseconds.

Flexible

Unlike traditional computer vision approaches, our models learn to extract discriminative features from the input using the provided training data, instead of using hand-engineered feature extractors like SIFT and LBP. This makes them easy to adapt to problems in any domain.

Diversity in AI is essential

It’s not so much about good vs bad AI. It’s about smart vs. dumb AI. As in, does this AI provide humans some value? Similarly, AI should be ethical, not to take people’s jobs but augment them by giving them 10x less work or make their work 10x faster.

If you only teach AI with a few subsets of the world, it will just understand these subsets to be the world. We at Clarifai have a clear stance on our responsibility to teach AI in a diverse and global way and provide solutions that don’t compete with your business and its values.

How does Clarifai’s technology compare to open source technology?

 

At Clarifai, we have been building CORE and optimizing our neural network architecture since 2013, when we took the top 5 spots at the ImageNet Large Scale Visual Recognition competition.

 

The level of accuracy that we have been able to achieve is a result of machine learning experts dedicated to optimizing the network architecture further, continually collecting high-quality data. Our models automatically improve their accuracy over time, so that’s one less thing your business has to worry about.

 

Don’t take it from us, see what our customers have to say.