Deci Introduces World’s Most Advanced Semantic Segmentation Models – EEJournal


Deci’s deep learning NAS technology has automatically generated a new model family called DeciSeg that offers unprecedented inference performance and accuracy

Tel Aviv, Israel, September 22, 2022]– Deci, the deep learning company that uses AI to build AI, today announced a new suite of industry-leading semantic segmentation models called DeciSeg. Deci’s proprietary Automated Neural Architecture Construction (AutoNAC) technology automatically generated semantic segmentation models that significantly outperform the best performing publicly available models, such as the Apple-released MobileViT and the Google-released DeepLab family. Deci models offer more than 2x lower latency and 3-7% higher accuracy.

Semantic segmentation is one of the most commonly used computer vision tasks in many industries including automotive, smart cities, healthcare, and consumer applications, and is commonly required for many edge AI applications. However, there are significant barriers to running semantic segmentation models directly on edge devices, such as B. High latency and the inability to deploy these models due to their size.

READ:  SmartCow Releases AI SuperCam and Box PC Powered by NVIDIA Jetson AGX Orin for Next-Generation Edge Applications – EEJournal

Semantic segmentation tasks that previously could not be performed at the edge because they were too resource-intensive are now possible with DeciSeg models. This enables companies to develop new use cases and applications on edge devices, reduce inference costs (since AI practitioners no longer have to perform these tasks in expensive cloud environments), enter new markets and reduce development times.

“DeciSegs exemplify the power of Deci’s AutoNAC engine to generate custom hardware-aware deep learning models with unprecedented performance on any hardware. AI teams can easily use DeciSegs models or leverage Deci’s AutoNAC engine to build and deploy custom models that run real-time computer vision tasks on their edge devices,” said Yonatan Geifman, PhD, co-founder and CEO of Deci.

Deci’s platform has a proven track record of enabling AI at the edge and empowering AI teams to build and deploy production-quality deep learning models. Earlier this year, Deci announced the discovery of DeciNets for CPUs, narrowing the gap between a model’s inference performance on a GPU and a CPU by halfwithout sacrificing the accuracy of the model, allowing the AI ​​to run on less expensive, resource-constrained hardware.

READ:  Upcoming Vasil Upgrade Earns #1 Spot for Cardano—The Top Coins to Watch for Sep 19 - Sep 25

“In the world of automated design and construction of deep neural networks, Deci’s AutoNAC technology is a game changer. It uses deep learning to search vast neural network spaces for the model that works best for a given task and AI chip. In this case, AutoNAC was applied to the Pascal VOC Semantic Segmentation Task on NVIDIA’s Jetson Xavier NX™ chip and we are very pleased with the results,” said Ran El-Yaniv, Co-Founder and Chief Scientist of Deci and Professor of Computer Science at Technion – Israel Institute of Technology.

Deci’s platform serves customers across multiple industries across multiple production environments including edge, mobile, data center and cloud. To learn more about how leading AI teams are using Deci’s platform to build production-quality models and accelerate inference performance, visit here.

READ:  Nvidia CEO Jensen Huang announces 'Hopper' GPU availability, cloud service for large AI language models

About Deci

Deci enables deep learning to reach its true potential by using AI to build better AI. The company’s deep learning development platform enables AI developers to build, optimize, and deploy faster, more accurate models for any environment, including cloud, edge, and mobile, enabling them to revolutionize industries with innovative products. The platform is built on Deci’s proprietary Automated Neural Architecture Construction (AutoNAC) technology, which automatically generates and optimizes the architecture of deep learning models, enabling teams to accelerate inference performance, enable new use cases on limited hardware, accelerate development cycles shorten and reduce computational costs. Founded by Yonatan Geifman, Jonathan Elial, and Professor Ran El-Yaniv, the team of deep learning engineers and scientists is dedicated to removing production-related bottlenecks throughout the AI ​​lifecycle.





Source link