Computer Vision is a field in which computers gain a high-level understanding of images or videos. It is an interdisciplinary field that is broadly a subfield of Artificial Intelligence.
It was meant to mimic the human visual system to provide robots with intelligence. It involves the process of acquiring, processing and analyzing images to extract features from the image better. Neural Networks are mostly used in the field of computer vision for processing due to the advantage of neural networks over machine learning algorithms as mentioned in my previous post
A few tasks based on Computer Vision are:
- Object Recognition
- Pose Estimation
- Character Recognition
- Scene reconstruction
- Image restoration
Object Recognition:
It is used for identifying objects in images and videos. It is widely used in video surveillance and person identification. The output is usually a bounding box representing the object in the image or segmented objects.
Pose Estimation:
It is problem in computer vision to detect the posture and orientation of object generally humans. These are used in the area of augmented reality and computer graphics.
Image Restoration:
It is the process of restoring the damaged and archaic, pixelated images. This application is applicable because of GAN (Generative Adversarial Networks).
Application:
- Computer vision is used in the medical field to diagnose patients using x-ray and other imaging methods. Example: tumor detection.
- Another booming area in computer vision is the development of autonomous vehicles which provide vision based path planning and movement
- Computer vision is also used in many industries for auditing and inspection of mechanical machines
These are the few examples that use computer vision there are lot many areas such fashion technology for recommendations on styles and storage based companies such as google photos use them improve the quality of the image.
