welcome to

Qualcomm
Innovation Center

open source, wireless, open software, innovation, Python, TensorFlow, JavaScript, Typescript

 

Empowering Users to Innovate

Open source and community-driven software development is becoming increasingly important to the wireless industry. The world is becoming more and more connected, from intelligent homes to autonomous vehicles and smart cities. Open source software is helping spur the creativity and innovation that will allow these devices to better support their users and provide even greater value. In recognition of this market dynamic, Qualcomm formed the the Qualcomm Innovation Center (QuIC).


Visit QuIC on GitHub
Brain on a chip image

What is QuIC?

Qualcomm Innovation Center is a wholly-owned subsidiary of Qualcomm Technologies, Inc., focused on enabling and optimizing open-source software for use with Qualcomm technology.

The company is focused on technical innovations to mobile open-source software that leverages the unique capabilities of today's mobile devices, with a particular eye on optimizing platforms and applications which run on them via tight hardware integration.

Projects

AI Model Efficiency Toolkit

Qualcomm Innovation Center (QuIC) is at the forefront of enabling low-power inference at the edge through its pioneering model-efficiency research. QuIC has a mission to help migrate the ecosystem toward fixed-point inference. With this goal, QuIC presents the AI Model Efficiency Toolkit (AIMET) - a library that provides advanced quantization and compression techniques for trained neural network models. AIMET enables neural networks to run more efficiently on fixed-point AI hardware accelerators.

About Repo

AIMET Model Zoo

The AIMET Model Zoo is a collection of popular pre-trained models optimized for 8-bit inference. It also provides the recipe for quantizing popular 32-bit floating point (FP32) models to 8-bit integer (INT8) models with little loss in accuracy. The tested and verified recipes include a script that optimizes TensorFlow or PyTorch models across a broad range of categories from image classification, object detection, semantic segmentation, and pose estimation to super resolution, and speech recognition.

About Repo

Sample Apps for Qualcomm Robotics RB5 Platform

Sample applications of Qualcomm's Robotics development platform demonstrating superior image capture and processing with depth mapping. Use of Qualcomm's AI engine achieves high-speed and high-performance on-device machine learning, as well as real-time object detection and pose estimation, hardware accelerated video encoding/decoding and much more, all within reasonable thermal limits.

Repo