The head background
CAISA
  • 90% Theoretical Peak Performance 90% Theoretical Peak Performance
  • 3x Increase in Per-Unit Computing Power 3x Increase in Per-Unit Computing Power
  • 5ms Latency 5ms Latency
Nebula Accelerator

Nebula Accelerator

Providing deep-learning-based object detection,tracking, recognition and attibute analysis intergrated back-end solution.

Processing Capacity

16-channel 1080p video streams

Full-scale real-time object detection and video

structuralization analysis with minimum object

size of 60x60 pixels

backend Read more
Rainman Accelerator

Rainman Accelerator

Providing deep learning-based object detection and tracking intergrated front-end solution.

Processing Capacity

1080p video stream

60x60 pixel specific target full detection

frontend Read more

Advantages of "Nebula" and "Rainman"

  • High
    performance

    High performance

    200FPS,

    16-channel real-time detection of single card

  • Low
    latency

    Low latency

    10 ms

  • Low power
    consumption

    Low power consumption

    CAISA architecture,

    10 times energy efficiency ratio Increased

Low barrier and full coverage application

  • Image recognition
  • Semantic segmentation
  • Time series analysis
  • Speech semantics