EPROLO

ScrollGuard for iOS is an app designed to limit endless scrolling on social media platforms. It helps users block distractions from Instagram Reels, YouTube Shorts, and TikTok. ScrollGuard aims to help users maintain social connections while reducing time spent on algorithm-driven content. It is free to download with core blocking features available without payment.

MobileVLM is an open-source vision language model designed for mobile devices. It enables on-device AI processing, offering a faster and stronger baseline for vision language tasks. MobileVLM facilitates the development of mobile AI assistants and applications requiring visual understanding.

TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi is a tutorial for training, converting, and running TensorFlow Lite object detection models. It supports deployment on Android devices and the Raspberry Pi. The tutorial provides step-by-step instructions for training models and deploying them on edge devices.

android-speech is an open-source library for Android speech recognition and text-to-speech. It simplifies the integration of voice capabilities into Android applications. Developers can easily add speech recognition and text-to-speech functionality. The library is available on GitHub.

LLMFarm is an iOS and MacOS application that allows users to work with large language models (LLMs) offline. It uses the GGML library to load and run different LLMs with specific parameters. LLMFarm enables users to test the performance of various LLMs on their Apple devices.

DeepBeliefSDK is a framework for image recognition. It implements a convolutional neural network architecture optimized for mobile devices. The SDK supports iOS, Android, Linux, and OS X platforms. It is designed to function within the memory and processing constraints of modern mobile devices.

uno is an open-source platform for building cross-platform native applications. It supports mobile, web, desktop, and embedded apps. Developers can create single-codebase apps using C#/XAML from any IDE. Hot Reload is included.

mace is a deep learning inference framework optimized for mobile heterogeneous computing platforms. It supports deployment on Android, iOS, and Linux. The framework is designed to optimize performance of AI models on mobile devices. It is an open-source project.

PoseEstimationForMobile is an open-source project providing real-time single-person pose estimation for Android and iOS. It implements CPM and Hourglass models using TensorFlow, utilizing inverted residuals for efficient inference. The tool enables pose estimation in mobile applications with optimized performance.