To interact with us robots need to see, hear, speak, and express themselves as naturally as we can. Furhat is specifically designed for human social interactions using a multimodal system that allows it to interact with us just as we interact with each other.
Furhat's swapable polymer magnetic mask is designed with adult human curvatures to display human facial elements of any size and shape, including eyebrows, eyes, nose, mouth, lips and more. Child and Anime masks are also available on demand.
Furhat uses a back-projection system for real time high-fidelity rendering of multi-layered textures that can be customized to create a life-like and dynamic face of any age, ethnicity and gender.
Furhat's visual perception system is comprised of a 120 degree diagonal field-of-view camera that includes a highly accurate face detection engine based on state-of-the-art deep learning (single-shot-detector) model, depth estimation, and spatial modelling.
Built-in microphones and an external USB multi-array microphone with noise cancelation and direction of arrival for audio localization. High fidelity speakers with volume control that give Furhat a clear and distinguishable voice in noisy environments.
Furhat's LED ring enables a silver lining effect, giving the robot the ability to signal when events occur in the system, further increasing its presence in a space when needed.
Furhat comes with several ports for input and output, including an ethernet port, audio jack and USB ports for connecting devices such as microphones, headphones, keyboards, and more.
Furhat comes with an automatic lip-syncing model and gestures such as blinking, smiling, nodding, frowning, and other gestures that can be controlled individually or fully automated with Large Language Models.
The groundbreaking combination of back-projected facial animations, replaceable masks and face customization tools make it easy to create expressive characters of any age, ethnicity and gender.
Furhat uses computer vision to track up to 10 multiple users in real-time, estimate user engagement and distance with the ability to distinguish between them and remember them during the course of an interaction. You can also easily stream your video feed to external models for more advanced user and environment analysis.
Build applications from scratch with the Furhat SDK or fully automate them with our FurhatAI Creator tool.
Want to get a detailed overview of Furhat's modular system? View or download the technical specifications below.