The Computer Vision Revolution
Computer vision is a field within AI that gives machines a human-like ability to see and understand the world. This cutting-edge technology is transforming lives by enabling a wide range of devices to extract meaningful data from digital images and video.
Real-World Use Cases
Developers and data scientists across industries use Intel's computer vision and AI technology to build solutions that solve some of the world's toughest challenges.
The latest computer vision technologies are transforming medical imaging by streamlining workflows to help improve patient care with faster and more accurate diagnosis.
Computer vision can help retailers understand where to place products, when to restock inventory, and who their customers are through demographics.
Smart city technologies can help gather video feeds from street cameras so city leaders can make more informed operational decisions.
How Computer Vision Works
Computer vision mimics how the visual cortex of the human brain works. Like our brains, it can recognize and classify what it sees and even infer information from new data. But first, it must be trained using a type of AI called deep learning.
Training a computer vision application requires inputting large amounts of data into a use case-specific deep learning model. The more data that the model analyzes, the better the application is with recognizing what it has been developed to identify. For example, for a computer vision application to recognize a bird, it must first learn what a bird looks like. It does this by analyzing thousands, maybe even millions, of images of birds of different sizes, colors, and characteristics.
Equipped with this information, the application has the knowledge it needs to interpret and infer information from digital images.
Convolutional Neural Networks
Deep learning models used in computer vision are from a family of algorithms known as convolutional neural networks (CNNs). These networks analyze the RGB values embedded in pixels to detect identifiable patterns. CNNs can be developed to evaluate pixels based on a wide range of features that include color distribution, shape, texture, and depth, and accurately recognize and classify objects.
Input
The process of introducing labeled images into a deep learning model.
Train
The model analyzes the images to build a database of application-specific knowledge.
Interpret
The trained model can recognize and classify specific visual information.
Infer
The model uses inference to recognize and classify new visual information.
Using Computer Vision to Perform a Wild Elephant Census
Computer vision has recently become a critical tool for wildlife conservation. In this example, data scientists used the technology to conduct a census of wild elephants. By inputting drone-captured video into a specially developed deep learning model, the scientists were able to quickly and accurately determine the size of the region's elephant population.
01
Object Recognition
Deep learning models must be trained to recognize specific objects. In this example, data scientists inputted thousands of labeled elephant images so the models could learn to identify the animal in the wild.
02
Image Classification
Trained deep learning models enabled the census application to differentiate between elephants and antelopes and correctly categorize each animal.
03
Object Detection
When the deep learning models detected an elephant, they highlighted the animal's position with a color-coded bounding box labeled with the appropriate classification.
04
Image Segmentation
Using color to distinguish pixels that form the image of the elephant, scientists gained more detailed information about the animal's position.
Intel's Computer Vision Portfolio
Intel offers an end-to-end AI software portfolio for use cases across computer vision, natural language processing, audio, and recommender systems.
Intel® Distribution of OpenVINO™ Toolkit
This free, open source toolkit helps AI inferencing models run faster and makes them easier to deploy on your existing hardware. The OpenVINO™ toolkit was founded in computer vision—it's even in the name: Open Visual Inference and Neural Network Optimization.
Computer vision applications across industries can benefit from the performance enhancements provided by the OpenVINO toolkit, powered by oneAPI.
With a public or pretrained model, C++ and Python* developers can write once and deploy anywhere to continue their workflow using common frameworks. These frameworks include TensorFlow*, PyTorch*, and ONNX* (Open Neural Network Exchange).
Intel® Geti™ Software Platform
This software platform enables enterprise teams to rapidly build computer vision AI models for manufacturing, retail, logistics, and many other industries.
Through an intuitive graphical interface, you can:
- Add images or video data.
- Make annotations.
- Train, retrain, export, and optimize AI models for deployment.
Data scientists, AI professionals, and industry experts can build production-ready AI models quickly and collaboratively in a single interface.
Equipped with powerful technology that includes active learning, task chaining, and smart annotations, the platform empowers teams to innovate with AI.
Recommended Resources
Hard Iron Meets AI at John Deere*
This manufacturer uses Intel® AI Technology to help solve a costly, age-old problem in the manufacturing welding process.
Power the Future with Intel® Technology
Technology is helping transform industries around the world. For example, as more industries modernize with leading-edge technology, Intel works with these customers to provide solutions that reduce costs and boost productivity.