AI Benchmark Quantifies Which Android Runs the Best Internals
Justin Lam — July 27, 2018 — Tech
References: play.google & venturebeat
Designed by researchers at ETH Zurich, AI Benchmark is a new Android-specific app that was developed to evaluate the performance of various smartphones. This testing is done across a suite of open-source algorithms that perform facial recognition, image super-resolution, image classification, photo enhancement, segmentation and a host of others to measure performance. AI Benchmark is even capable of testing performance on neural networks used in driverless cars, which may one day be run on chips comparable to the ones found in smartphones.
As AI Benchmark runs the phone through its paces, it also produces visualizations of the phone's performance. This numeric score factors in the speed of the system-on-chip and available RAM to produce an accurate and pertinent score of the device.
As AI Benchmark runs the phone through its paces, it also produces visualizations of the phone's performance. This numeric score factors in the speed of the system-on-chip and available RAM to produce an accurate and pertinent score of the device.
Trend Themes
1. Benchmark-testing Apps - Developing new benchmark-testing apps that can accurately evaluate the performance of smartphones and other devices.
2. Facial Recognition - Advancing facial recognition technology to improve accuracy and speed in various applications such as security, authentication, and personalization.
3. Neural Network Performance - Enhancing the performance of neural networks to ensure optimal functionality in driverless cars and other AI-driven applications.
Industry Implications
1. Technology - Opportunities for technology companies to develop and offer benchmark-testing apps to assess the performance of smartphones and other devices.
2. Artificial Intelligence - Disruptive innovation opportunities to improve facial recognition algorithms and neural network performance for a wide range of industries.
3. Automotive - Exploring the use of AI benchmark-testing apps to optimize the performance of neural networks and chips in driverless cars for improved safety and efficiency.
0.8
Score
Popularity
Activity
Freshness
