When a user takes a photo or video, these technologies enable his or her smart device to understand the context of the subject within the environment. It auto-detects that environment holistically, analyzing the what, the when and the where – for example, whether the image is taken of a person, a pet, or any number of subjects, and whether it is at the beach, in a snowscape, on a city street, or a rural environment during a specific time of day such as nighttime, daytime, and so on.
At the moment of capture, this technology automatically adjusts the parameters in order to optimize the quality of the image based on the subject and environmental factors, as mentioned above. Post-capture, the user can more readily search or index the contents based on the environment in which it was taken.
We've assembled a large database of numerous indoor and outdoor environments and objects that allow us to recognize and more accurately identify and analyze various scenes. Scene detection also allows cameras to automatically select optimal settings for photo-taking across various environments.
Pictures and video can be categorized by color, shape, scene and other qualities, allowing for easier search and indexing of detailed image contents.
Following the continuous advancement of smart technologies into people’s lives and the workplace, the camera has been placed at the forefront of this movement due to its extremely important role as the “eye”. ArcSoft object recognition technology takes this movement one step further, essentially functioning as the “brain” through the detection and recognition of all visible objects within an image or video, thus playing a crucial part in the enhancement of smart technologies.
Within images and videos, objects tend to go through numerous changes including differences in size, positioning, or partial obstruction. Through the use of deep learning by analyzing enormous quantities of various objects in different situations, ArcSoft has developed technology that can accurately detect a wide range of objects with different sizes or dimensions. Object recognition technology has already met a variety of applications in all areas, from an oven recognizing food, cars recognizing humans, other cars, or roads, and robots that can recognize the layout of a room, all the way to augmented reality (AR) and virtual reality (VR) devices that can recognize objects ahead of the user. Object recognition technology plays a vital role in all areas in today’s world.