Warm Up Your Smartphone's Neurons with Deep Learning
summary
A team from Xebia presents a mobile-focused proof of concept that uses deep learning for on-device object recognition. They cover Android Neural Networks API, TensorFlow Lite, and Apple CoreML, and discuss the impact of these models on smartphone performance, cloud training and deployment architecture, and building the mobile app using the latest platform features.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 3:35Theory to practice
- 6:24Neural networks demystified
- 10:14Training and transfer learning
- 13:24Android optimization limits
- 17:40MobileNets and model reduction
- 20:50Android app integration
- 23:40Deployment pipeline and cloud
- 27:56Benchmarking performance
- 30:28Live app demo
- 33:18Future trends
- 35:10Q&A
related talks
20mHeat the neurons of your Android smartphone with Deep Learning
46mFire Up Your Smartphone Neurons with On-Device Deep Learning
39mAI/ML on mobile/android with TensorFlow
29mRunning Deep Learning Models on Mobile with optimised speed
15mSimplified Machine Learning on mobile, browsers, and edge devices with on-device ML
47mDeep Learning: a practical introduction from an Android dev perspective
39mMachine Learning on Android Demystified
38mNEURAL NETWORKS ON YOUR MOBILE PHONE. DIVING INTO THE NN API