Overview

Description

Founded in 2014, STRADVISION is an automotive industry pioneer in deep learning-based vision perception technology for ADAS. STRADVISION’s automotive-grade custom software SVNet is being deployed on various vehicle models in partnership with OEMs. In addition, SVNet has achieved TISAX’s AL3 standard for information security management, as well as being certified to the ISO 9001:2015 for Quality Management Systems and ISO 26262 for Automotive Functional Safety. STRADVISION has surpassed 1 million units in cumulative commercial production since 2019.

STRADVISION's '3D Perception Network,' a revolutionary advancement in autonomous systems, elevates the camera's role in Advanced Driver Assistance Systems (ADAS) and Autonomous Driving (AD), offering a comprehensive solution from perception to planning. The traditional sensing-perceiving-planning-action sequence undergoes a pivotal evolution, driven by deep learning for a refined 2D to 3D transformation.

Developed by STRADVISION, the 3D Perception Network represents a significant breakthrough in the ADAS/AD field. This approach leverages LiDAR sensors during development but distinguishes itself by transitioning into a robust and cost-effective alternative, eliminating the need for in-vehicle LiDAR systems. This strategic decision underscores STRADVISION's commitment to cutting-edge technology and its practical application in ADAS/AD systems, reducing implementation costs and enhancing scalability in the automotive industry.

Key Highlights:

  • Higher Scalability: The 3D Perception Network achieves unmatched scalability, with a 30% decrease in ARM resource usage. This innovation supports flexibility across various SoC platforms, optimizing efficiency for the Software Defined Vehicle (SDV) era.
  • Less Complexity: Implementation of the 3D Perception Network reduces post-processing code complexity by 50%, streamlining integration and maintenance processes.
  • Expansion of Camera Role: Redefining the role of the camera, the 3D Perception Network competes with Lidar capabilities, marking a significant leap in vision's role in autonomous systems.
  • Enhanced Cost Efficiency: Offering advancements in scalability and reduced complexity, the 3D Perception Network significantly enhances cost efficiency, making it an attractive choice for OEMs integrating cutting-edge ADAS/AD technologies.
  • Better Accuracy: This innovative approach ensures superior accuracy, reducing detection errors by 10% and boosting detection stability by 30%, positioning STRADVISION as the go-to choice for OEMs seeking precision and dependability in autonomous driving solutions.

Features

  • Road Object
    • Road ObjectVehicle Detection & Recognition
    • Pedestrian Detection & Recognition
    • Two-wheeler Detection & Recognition
    • Light Source Detection & Recognition
    • Traffic Sign Detection & Recognition
    • Traffic Light Detection & Recognition
    • General Object Detection
    • Static Object Detection & Recognition
  • Road Surface
    • Lane Detection & Recognition
    • Road Boundary Detection & Recognition
    • Road Marking Detection & Recognition
    • Free Space Detection
    • Road Elevation Profile
    • Road Slope Estimation
    • Parking Space Detection
  • Environment
    • Scene Classification
    • Semantic Segmentation
  • Geometry
    • DepthMap
    • Visual Odometry
    • Visual Positioning
    • Sensor Fusion
    • Visualization

Target Devices

Videos & Training

STRADVISION’s SVNet demo on RENESAS Platform (R-Car V3H2)

SVnet satisfies EU GSR and Euro NCAP5 star aiming at ADAS L0-L2.