“VisualCortex” is the application responsible for the visual capabilities of Human-like AI platform “LUDENS”. Spatial information is mapped out like human vision to include movement, posture, facial expression, gaze, line of sight, and more. These characteristics are analyzed in real-time from the aggregated area.
The visually specialized AI technology behind VisualCortex has achieved various results around the world.
Sponsored by IEEE
AI Competition
Sponsored by Facebook Research
VR/AR Recognition Accuracy Competition
Computer Vision
International Conference CVPR 2019 workshop
Detects people and their locations in the image
Track the same person across multiple frames and multiple people
Detects the face and position of a person in an image
Detects a person's gaze and recognizes whether the person is gazing or not
Estimates emotions from detected human facial expressions
Detects whether or not someone has a mask on
Detects the direction of a person's gaze
A stand-alone, simple, inexpensive system that emphasizes speed and can be up and running immediately.
We recommend using the Jetson Nano™ module, a system-on-module that enables the development of small, low-cost, low-power AI devices. VisualCortex is designed for AI implementation in a variety of industries from smart cities to robotics. Reference: Jetson Nano™ official website
VisualCortex is designed to work with common webcams available on the market. There is no need to purchase a special, dedicated camera.
Recognition results can be obtained in JSON (JavaScript Object Notation). It is a lightweight text-based data exchange format that can be used to link data to various external systems. It is also used in "LUDENS," a Human-like AI developed by Cougar. JSON Official Website
Detects people and their positions in an image
Output | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Recognition probability | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Position |
Detects and tracks the same person in the video
Output | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Person ID |
Detects the face and position of a person in an image
Output | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Recognition probability | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Position |
Detects a person's line of sight and recognizes whether the person is gazing at it or not
Output | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Attention flag (0 or 1) |
Estimates and predicts emotions based on detected human facial expressions.
Output | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Recognition probability | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Emotion type |
|
Detects whether a mask is worn.
Output | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Recognition probability | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Position |