Products / IoT, Industrial & Robotics

Bringing artificial intelligence to the

AI-powered applications are often highly vision dependent, making them ideally suited for Ambarella chips. Our advanced stereovision processing transforms the environment into a fully 3-dimensional world with exquisite detail, enhancing both safety and performance, while our CVflow® architecture delivers real-time execution of neural networks in a variety of challenging conditions, including high-contrast or low-light scenes. We deliver highly accurate environmental modeling, low latency, and real-time processing for instant decision-making in a wide range of applications—controlling an autonomous robot, identifying a defect on a production line, tracking inventory, managing access to buildings, and much more. The result? AI applications that are simply smarter, faster, safer, and more efficient than ever before.

CV5S

Ambarella’s CV5S provides 8K image processing, video encoding / decoding, and CVflow® computer vision processing in a single, low-power design. Fabricated in advanced 5 nm process technology, CV5S achieves power consumption below 4W for 8K video recording at 30 fps—suited for IP, robotics, and machine vision applications.

Key features

  • 8KP30 video encoding performance with multi-streaming capability
  • CNN toolkit for easy porting of neural networks implemented in Caffe, TensorFlow, PyTorch, or ONNX frameworks
  • Superior low-light processing

CV52S

Ambarella’s CV52S provides 4K image processing, video encoding / decoding, and CVflow® computer vision processing in a single, low-power design while achieving power consumption below 3W for 4KP60 video recording with advanced AI processing at 30 fps.

Key Features

  • 4KP60 encoding with multiple substreams video performance
  • CNN toolkit for easy porting of neural networks implemented in Caffe, TensorFlow, PyTorch, or ONNX frameworks
  • Enhanced secure boot with TrustZone® and secure memory, true random number generator (TRNG), one-time programmable memory (OTP), Arm trusted base system architecture (TBSA) compliance, DRAM scrambling, and virtualization
  • 4K IP cameras with advanced artificial intelligence (AI), intelligent traffic systems (ITS) cameras, machine vision and robotics applications

CV2S

Designed for high-performance surveillance cameras with advanced computer vision features, the CV2S chip delivers industry-leading image processing, stereovision capabilities, 4KP60 video encoding, along with CVflow® computer vision architecture.

Key features

  • CNN / DNN-based processing: detection, classification, tracking, and more
  • Dedicated hardware for stereo processing
  • Dedicated CVflow® computer vision processor
  • Tools for high and low-level algorithm development
  • CNN toolkit for easy porting with Caffe, TensorFlow, and ONNX
  • Support for global shutter sensors and high-frame-rate capture
  • Advanced image processing with HDR, dewarping, EIS, MCTF, and more
  • Secure boot with TrustZone® and secure memory, true random number generator (TRNG), one-time programmable memory (OTP), DRAM scrambling and virtualization

CV22S

Ambarella’s CV22S SoC combines advanced image processing, 4KP30+ video encoding, powerful CVflow® computer vision performance, and a full suite of cybersecurity features to enable the next generation of intelligent IP camera designs.

Key features

  • CNN / DNN-based processing: detection, classification, tracking, and more
  • Dedicated CVflow computer vision processor
  • Tools for high and low-level algorithm development
  • CNN toolkit for easy porting with Caffe, TensorFlow, and ONNX
  • High-efficiency video encoding up to 4KP30+ performance
  • Support for visual odometry and simultaneous location and mapping (SLAM) algorithm implementation
  • Advanced image processing with HDR, dewarping, EIS, MCTF, and more
  • Secure boot with TrustZone® and secure memory, true random number generator (TRNG), one-time programmable memory (OTP), DRAM scrambling and virtualization

CV25S

The CV25S chip powers the next generation of affordable and intelligent home monitoring and professional surveillance solutions on the edge, combining advanced image processing, high-resolution video encoding, CVflow® computer vision performance, and a full suite of cybersecurity features.

Key features

  • CNN / DNN-based processing: detection, classification, tracking, and more
  • Dedicated CVflow computer vision processor
  • Tools for high and low-level algorithm development
  • CNN toolkit for easy porting with Caffe, TensorFlow, and ONNX
  • Advanced image processing with HDR, dewarping, EIS, MCTF, superior low-light imaging, and more
  • High-efficiency video encoding up to 4KP30+ performance
  • Advanced image processing with HDR, dewarping, EIS, MCTF, and more
  • Secure boot with TrustZone® and secure memory, true random number generator (TRNG), one-time programmable memory (OTP), DRAM scrambling and virtualization

CV28M

Our CV28M SoC is designed for intelligent applications at the edge, including smart home security, retail monitoring, consumer robotics, and occupancy monitoring. The chip’s high-performance CVflow® architecture provides deep neural network (DNN) computer vision processing and 4KP30 video encoding, enabling an array of computer vision-based devices.

Key features

  • CVflow® computer vision engine
  • CNN / DNN inference acceleration for detection, classification, and more
  • CNN toolkit for easy porting with Caffe, TensorFlow, and ONNX
  • Up to 4KP30 video performance
  • Advanced image processing with HDR, dewarping, EIS, MCTF, superior low-light imaging, and more
  • Secure boot with TrustZone® and secure memory, TRNG, OTP, DRAM scrambling and virtualization