Deep Fusion AI
CEO. Sunghun Yu

Autonomous driving cost innovation via DFAI’s 4D imaging radar solution
#AI #Mobility #4D_imaging_radar_solutions
Company Info.
Main Product/Service. 4D imaging radar solution
Date of Establishment. February, 2022
Contact Info.
Company Profile
Deep Fusion AI is the first in the world to develop applications such as 360-degree deep learning and HD maps, which are essential for autonomous driving. It utilizes 4D imaging radar, which is expected to contribute to the widespread adoption of autonomous driving by replacing expensive LiDAR.
This technology can be implemented by OEMs and Tier 1 suppliers and is provided as a solution.
This will contribute to the development of not only autonomous vehicles, but ships, and transport robots. Deep Fusion AI introduces an innovative methodology for 4D imaging radar deep learning and demonstrates its effectiveness by installing it on target vehicles of several overseas automotive OEMs.
Product / Service
What makes us different
The 4D imaging radar's 360-degree deep learning technology, which Deep Fusion AI will introduce at CES 2025, offers several innovative features. This radar is a radio-based sensor and, unlike optical sensors like cameras or LiDAR, does not require a standardized dataset for deep learning.
By implementing simulations of 4D imaging radars according to each manufacturer's specifications in a virtual environment and developing a pre-training model for deep learning, we provide performance targets to 4D imaging radar manufacturers and prepare them on time for product launches. This allows deep learning applications to be used simultaneously. Additionally, we offer automobile manufacturers the ability to select a 4D imaging radar that fits the mounting location, price, and purpose appropriate for each vehicle.
Work Experience
2024
2023
2022
Oct / Korean defence contract signed
Oct / Chinese Automotive Tier1 contract signed
Jan / Eureka at CES 2024
Sep / Automotive Tier1 contract signed
Jun / Prototype Launching
Feb / Establishment