About NextVPU

NextVPU designs and manufactures embedded AI visual processing chips, such as the NE-D163A, which enable high-resolution depth data capture and processing for smart devices without the need for external processors. Their technology addresses the challenges of low accuracy and high latency in depth perception for applications in security, robotics, and consumer electronics.

```xml <problem> Many smart devices struggle with accurately capturing and processing high-resolution depth data due to limitations in processing power and the need for external processors, leading to increased latency and reduced accuracy in depth perception. This is particularly challenging for applications in security, robotics, and consumer electronics that require real-time depth information. </problem> <solution> NextVPU designs and manufactures embedded AI visual processing chips that enable high-resolution depth data capture and processing directly on smart devices, eliminating the need for external processors. Their chips, such as the NE-D163A, integrate a proprietary CVKit™ neural network core, offering high energy efficiency with up to 2.4TOPS/INT8 and 1.2TOPS/FP16 performance. This integration allows for real-time, high-accuracy depth mapping with low latency, even in challenging lighting conditions. NextVPU's solutions provide a complete hardware and software platform for accelerating computer vision tasks in edge computing environments. </solution> <features> - NE-D163A chip integrates ISP, DSP, CNN, and depth processing capabilities for high-resolution depth map generation. - CVKit™ neural network core provides industry-leading energy efficiency for AI computation. - Supports 4-channel 1080P encoding/decoding and ISP for multi-channel video processing. - Compliant with AEC-Q100 Grade 2 standards for automotive ADAS applications. - Feynman M1 smart depth camera utilizes active light binocular vision and the NE-D163A chip for high-precision depth data output via USB 3.0. - InferStudio™ software development tool for AI model deployment and optimization. </features> <target_audience> The primary target audience includes manufacturers of security systems, robotics, consumer electronics, and automotive ADAS systems seeking to integrate high-performance, low-latency depth sensing capabilities into their products. </target_audience> ```

What does NextVPU do?

Where is NextVPU located?

NextVPU is based in Shanghai, China.

Location

Shanghai, China

Employees

7 employees

Find Investable Startups and Competitors

Search thousands of startups using natural language

AI voice (2021+)underground pipe robots energy flexibility software

Start Searching

NextVPU

⚠️ AI-generated overview based on web search data – may contain errors, please verify information yourself! You can claim this account with your email domain to make edits.

Executive Summary

nextvpu.com 10+

Shanghai, China

Funding

No funding information available.

Team (5+)

No team information available.

Company Description

Problem

Many smart devices struggle with accurately capturing and processing high-resolution depth data due to limitations in processing power and the need for external processors, leading to increased latency and reduced accuracy in depth perception. This is particularly challenging for applications in security, robotics, and consumer electronics that require real-time depth information.

Solution

NextVPU designs and manufactures embedded AI visual processing chips that enable high-resolution depth data capture and processing directly on smart devices, eliminating the need for external processors. Their chips, such as the NE-D163A, integrate a proprietary CVKit™ neural network core, offering high energy efficiency with up to 2.4TOPS/INT8 and 1.2TOPS/FP16 performance. This integration allows for real-time, high-accuracy depth mapping with low latency, even in challenging lighting conditions. NextVPU's solutions provide a complete hardware and software platform for accelerating computer vision tasks in edge computing environments.

Features

NE-D163A chip integrates ISP, DSP, CNN, and depth processing capabilities for high-resolution depth map generation.

CVKit™ neural network core provides industry-leading energy efficiency for AI computation.

Supports 4-channel 1080P encoding/decoding and ISP for multi-channel video processing.

Compliant with AEC-Q100 Grade 2 standards for automotive ADAS applications.

Feynman M1 smart depth camera utilizes active light binocular vision and the NE-D163A chip for high-precision depth data output via USB 3.0.

InferStudio™ software development tool for AI model deployment and optimization.

Target Audience

The primary target audience includes manufacturers of security systems, robotics, consumer electronics, and automotive ADAS systems seeking to integrate high-performance, low-latency depth sensing capabilities into their products.