Resource processing breakdown
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.
Physical resource mapping
During the build flow, the Hailo dataflow compiler decomposes each network layer into the necessary computational elements. This process generates a resource graph that represents the target network.
Dynamic configuration and execution
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.
The Hailo-8™ AI accelerator brings industry-leading neural processing throughput and power efficiency to support a wide range of AI applications. The table below demonstrates some of Hailo-8’s best-in-class capabilities in the foundational neural tasks of object detection and classification. It achieves very high FPS and power efficiency (FPS/W) across a number of industry-standard neural network models, while also offloading post-processing from the host processor it is working with.
NN Model | Input Resolution | FPS | Power [W] | FPS/ W |
---|---|---|---|---|
Classification | ||||
ResNet-50 v1 | 224×224 | 1,332 | 3.45 | 386 |
MobileNet_v2_1.0 | 224×224 | 2,444 | 2.152 | 1,135 |
Object Detection | ||||
SSD_MobileNet_v1 | 300×300 | 1,055 | 2.2 | 479 |
YOLOv5m | 640×640 | 218 | 4.6 | 47.3 |
Segmentation | ||||
stdc1 | 1024×1920 | 54 | 2.9 | 18.6 |
Multi stream object detection (8 streams) | ||||
YOLOv3 | 608×608 | 69 | 4.9 | 14 |
Automotive
Security
Industrial Automation
Retail
Hailo offers breakthrough AI accelerators and Vision processors
Our Partners
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Eget felis eget nunc lobortis mattis aliquam faucibus purus. Egestas purus viverra accumsan in nisl nisi scelerisque eu ultrices. Nulla facilisi cras fermentum odio eu feugiat pretium nibh ipsum. Vitae et leo duis ut. Aliquam eleifend mi in nulla
posuere sollicitudin aliquam ultrices. Sit amet luctus venenatis lectus magna fringilla urna porttitor rhoncus. Convallis posuere morbi leo urna molestie at
elementum eu facilisis. Quis enim lobortis scelerisque fermentum dui faucibus in ornare. Senectus et netus et malesuada. Pellentesque pulvinar pellentesque habitant morbi tristique senectus et netus et.
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Eget felis eget nunc lobortis mattis aliquam faucibus purus. Egestas purus viverra accumsan in nisl nisi scelerisque eu ultrices. Nulla facilisi cras fermentum odio eu feugiat pretium nibh ipsum. Vitae et leo duis ut. Aliquam eleifend mi in nulla
posuere sollicitudin aliquam ultrices. Sit amet luctus venenatis lectus magna fringilla urna porttitor rhoncus. Convallis posuere morbi leo urna molestie at
elementum eu facilisis. Quis enim lobortis scelerisque fermentum dui faucibus in ornare. Senectus et netus et malesuada. Pellentesque pulvinar pellentesque habitant morbi tristique senectus et netus et.
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Sign up to our newsletter to stay up to date
The Hailo-15™ family of AI vision processors combines Hailo’s patented and field proven AI inferencing capabilities with advanced computer vision engines, enabling unprecedented video processing and analytics at a premium image quality.
The Hailo-8™ AI accelerator allows edge devices to run deep learning applications at full scale more efficiently, effectively, and sustainably, with an architecture that takes advantage of the core properties of neural networks.
We invite you to join our rapidly expanding ecosystem of world-class partners and become a part of our global community.
Hailo presents a unique technology, designed to enable AI applications across a broad range of industries.
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
Join us at Electronica China in Shanghai, China.
Discover how you can boost your video analytics systems with the best-performing Hailo-8™ AI Processor for accurate and high-performance processing in real-time within an embedded power envelope.
Visit the Renesas booth # 7.2D202 where you will be able to see a live demo of 360° High-End Object Detection for L2+/L3 Automated Driving.
We are looking forward to scheduling a meeting or demo with you.
Fill in the form to meet our team!
A wide ecosystem of partners to help you design a complete solution
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam,
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae. Integer eget aliquet nibh praesent tristique magna sit amet. Nibh nisl condimentum id venenatis a condimentum vitae sapien pellentesque. Pellentesque habitant morbi tristique senectus et netus et.
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae. Integer eget aliquet nibh praesent tristique magna sit amet. Nibh nisl condimentum id venenatis a condimentum vitae sapien pellentesque. Pellentesque habitant morbi tristique senectus et netus et.
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae. Integer eget aliquet nibh praesent tristique magna sit amet. Nibh nisl condimentum id venenatis a condimentum vitae sapien pellentesque. Pellentesque habitant morbi tristique senectus et netus et.
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae. Integer eget aliquet nibh praesent tristique magna sit amet. Nibh nisl condimentum id venenatis a condimentum vitae sapien pellentesque. Pellentesque habitant morbi tristique senectus et netus et.
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae. Integer eget aliquet nibh praesent tristique magna sit amet. Nibh nisl condimentum id venenatis a condimentum vitae sapien pellentesque. Pellentesque habitant morbi tristique senectus et netus et.
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae. Integer eget aliquet nibh praesent tristique magna sit amet. Nibh nisl condimentum id venenatis a condimentum vitae sapien pellentesque. Pellentesque habitant morbi tristique senectus et netus et.
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae. Integer eget aliquet nibh praesent tristique magna sit amet. Nibh nisl condimentum id venenatis a condimentum vitae sapien pellentesque. Pellentesque habitant morbi tristique senectus et netus et.
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae. Integer eget aliquet nibh praesent tristique magna sit amet. Nibh nisl condimentum id venenatis a condimentum vitae sapien pellentesque. Pellentesque habitant morbi tristique senectus et netus et.
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Innovative control scheme based on a combination of hardware and software reaching very low joules/operation with a high degree of flexibility
Innovative control scheme
Distributed memory fabric with purpose-built pipeline elements that allow very low-power memory access in neural network processing
Extremely efficient computational elements that can be applied variably, as needed
Dataflow-oriented interconnect adapts to the structure of the neural network and allows high resource utilization
Hailo Dataflow Compiler – full-stack software co-designed with the hardware architecture of the neural network processor, enabling efficient deployment of neural network models with seamless integration to existing frameworks
Lorem ipsum – full-stack software co-designed with the hardware architecture of the neural network processor, enabling efficient deployment of neural network models with seamless integration to existing frameworks
Dataflow-oriented interconnect adapts to the structure of the neural network and allows high resource utilization
Hailo Dataflow Compiler – full-stack software co-designed with the hardware architecture of the neural network processor, enabling efficient deployment of neural network models with seamless integration to existing frameworks
Distributed memory fabric with purpose-built pipeline
Lorem ipsum – full-stack software co-designed with the hardware architecture of the neural network processor, enabling efficient deployment of neural network models with seamless integration to existing frameworks
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Innovative control scheme based on a combination of hardware and software reaching very low joules/operation with a high degree of flexibility
Innovative control scheme
Distributed memory fabric with purpose-built pipeline elements that allow very low-power memory access in neural network processing
Extremely efficient computational elements that can be applied variably, as needed
Dataflow-oriented interconnect adapts to the structure of the neural network and allows high resource utilization
Hailo Dataflow Compiler – full-stack software co-designed with the hardware architecture of the neural network processor, enabling efficient deployment of neural network models with seamless integration to existing frameworks
Lorem ipsum – full-stack software co-designed with the hardware architecture of the neural network processor, enabling efficient deployment of neural network models with seamless integration to existing frameworks
Dataflow-oriented interconnect adapts to the structure of the neural network and allows high resource utilization
Hailo Dataflow Compiler – full-stack software co-designed with the hardware architecture of the neural network processor, enabling efficient deployment of neural network models with seamless integration to existing frameworks
Distributed memory fabric with purpose-built pipeline
Lorem ipsum – full-stack software co-designed with the hardware architecture of the neural network processor, enabling efficient deployment of neural network models with seamless integration to existing frameworks
Amet risus nullam eget felis eget nunc lobortis mattis aliquam. In iaculis nunc sed augue lacus viverra vitae
Innovative control scheme based on a combination of hardware and software reaching very low joules/operation with a high degree of flexibility
Innovative control scheme
Distributed memory fabric with purpose-built pipeline elements that allow very low-power memory access in neural network processing
Extremely efficient computational elements that can be applied variably, as needed
Dataflow-oriented interconnect adapts to the structure of the neural network and allows high resource utilization
Hailo Dataflow Compiler – full-stack software co-designed with the hardware architecture of the neural network processor, enabling efficient deployment of neural network models with seamless integration to existing frameworks
Lorem ipsum – full-stack software co-designed with the hardware architecture of the neural network processor, enabling efficient deployment of neural network models with seamless integration to existing frameworks
The Model Explorer is intended for users who are in the process of selecting which Neural Network (NN) architecture to use on the Hailo AI processors. Selecting an appropriate model to use in your application can be hard. There are many considerations like inference speed, model availability, desired accuracy, license, and more. Inference speed is unique since it cannot be easily estimated without the underlying hardware used.
The Model Explorer is intended for users who are in the process of selecting which Neural Network (NN) architecture to use on the Hailo AI processors. Selecting an appropriate model to use in your application can be hard. There are many considerations like inference speed, model availability, desired accuracy, license, and more. Inference speed is unique since it cannot be easily estimated without the underlying hardware used.
Notes:
Hailo’s revolutionary architecture is a clean-slate approach to the design of a specialized technology stack. It has created a domain-specific processor that significantly outperforms the Von Neumann architecture for deep learning tasks.
Artificial Intelligence and Deep Learning are revolutionary technologies that transform the way we use machines to perceive and analyze the world around us and respond in real-time to constantly evolving environments.
Back in 2017, when we founded Hailo, those disruptive technologies were limited to data centers, as they are costly, require high compute power and extensive hardware, and consume a significant amount of power.
Artificial Intelligence and Deep Learning are revolutionary technologies that transform the way we use machines to perceive and analyze the world around us and respond in real-time to constantly evolving environments.
Back in 2017, when we founded Hailo, those disruptive technologies were limited to data centers, as they are costly, require high compute power and extensive hardware, and consume a significant amount of power.
Detecting and classifying objects within an image is a crucial task in computer vision, known as object detection. Deep learning models trained on the COCO dataset, which is a popular dataset for object detection, offer varying tradeoffs between performance and accuracy. For instance, by running inference on Hailo-8, the YOLOv5m model achieves 218 FPS and 42.46mAP accuracy, while the SSD-MobileNet-v1 model attains 1055 FPS and 23.17mAP accuracy. The COCO dataset includes 80 unique classes of objects for general usage scenarios, including both indoor and outdoor scenes.
Detecting and classifying objects within an image is a crucial task in computer vision, known as object detection. Deep learning models trained on the COCO dataset, which is a popular dataset for object detection, offer varying tradeoffs between performance and accuracy. For instance, by running inference on Hailo-8, the YOLOv5m model achieves 218 FPS and 42.46mAP accuracy, while the SSD-MobileNet-v1 model attains 1055 FPS and 23.17mAP accuracy. The COCO dataset includes 80 unique classes of objects for general usage scenarios, including both indoor and outdoor scenes.
Model Build Computer
Model Zoo, a variety of common and state-of-the-art pre-trained models and tasks in TensorFlow and ONNX.Hailo Dataflow Compiler, for offline compilation and optimization of user’s models for Hailo devices
Host processor
Hailo TAPPAS, set of full application examples, implementing pipeline elements and pre-trained AI tasksHailoRT production-grade and light runtime software package, running on the host processor for real-time inferencing the deep learning models compiled by the Dataflow Compiler
Hailo-15™
Hailo TAPPAS, set of full application examples, implementing pipeline elements and pre-trained AI tasksHailoRT production-grade and light runtime software package, running on the host processor for real-time inferencing the deep learning models compiled by the Dataflow Compiler 2
General Manager Europe
Or Danon, Founder and CEO of Hailo, presents the “Lessons Learned from the Deployment of Deep Learning Applications In Edge Devices” tutorial at the September 2020 Embedded Vision Summit.
October 6, 2022
When choosing an object detection network for edge devices, there are many factors you should consider: compute power, memory resources, and many more. Which ones? This blogpost outlines everything you need to know when choosing an object detection network for your edge application. But remember, the Hailo-8™ processor provides high-performance computing on the edge and can have a prominent role in improving the accuracy of the network.