Add offline, GPU-accelerated AI upscaling for images

Hi everyone, I wanted to share a project I have been working on to bring high-quality super-resolution models directly to Android devices without relying on cloud processing. I have developed RendrFlow, a complete AI image utility belt designed to perform heavy processing entirely on-device. The Tech Stack (Under the Hood): Instead of relying on an internet connection, the app runs the inference locally. I have implemented a few specific features to manage the load: - Hardware Acceleration: You can toggle between CPU, GPU, and a specific "GPU Burst" mode to maximize throughput for heavier models. - The Models: It supports 2x, 4x, and even 16x Super-Resolution upscaling using High and Ultra quality models. - Privacy: Because there is no backend server, it works in Airplane mode. Your photos never leave your device. Full Feature List: I did not want it to just be a tech demo, so I added the utilities needed for a real workflow: - AI Upscaler: Clean up low-res images with up to 16x magnification. - Image Enhancer: A general fix-it mode for sharpening and de-blurring without changing resolution. - Smart Editor: Includes an offline AI Background Remover and a Magic Eraser to wipe unwanted objects. - Batch Converter: Select multiple images at once to convert between formats (JPEG, PNG, WEBP) or compile them into a PDF. - Resolution Control: Manually resize images to specific dimensions if you do not need AI upscaling. Why I need your help: Running 16x models on a phone is heavy. I am looking for feedback on how the "GPU Burst" mode handles heat management on different chipsets . https://play.google.com/store/apps/details?id=com.saif.example.imageupscaler