MIRNet-TFLite-TRT
by sayakpaul
TensorFlow Lite models for MIRNet for low-light image enhancement.
AI summary
Image enhancer
Converts MIRNet model into TensorFlow Lite and TensorRT models for image enhancement
- stars
- 130
- forks
- 25
- watching
- 6
Similar projects
Found by comparing what the projects do, not just their names.
Model Profiler
An app for profiling and optimizing the performance of TensorFlow Lite models on mobile devices
Transformer model converter
Converts popular transformer models to run on Android devices for efficient inference and generation tasks.
Image enhancer
An implementation of a generative adversarial network (GAN) used to improve image quality through super-resolution
jakeret/tf_unet1.9K
Image segmenter
An open source U-Net image segmentation implementation using Tensorflow
Mobile ML tutorials
Tutorials and projects on building, deploying, and using TensorFlow Lite models for mobile applications
Squeeze Network
A TensorFlow implementation of Squeeze and Excitation Networks for image classification tasks
Object Detector
An implementation of YOLOv5 object detection model in TensorFlow Lite format
OCR model converter
A repository to convert popular OCR models into TensorFlow Lite (TFLite) formats for efficient deployment on mobile devices.
Segmentation model
Deep learning models for semantic segmentation of images
Image classifier
An Expo-based React Native application using TensorFlow.js to detect objects in images with MobileNet
ML Android apps
An Android app repository showcasing various machine learning models and their implementations in Kotlin
ML example repository
Refactoring and rewriting official TensorFlow Lite examples for Android to improve quality and readability
ML model deployment tool
A toolkit to deploy machine learning models on mobile devices using the TensorFlow Lite framework
Game controller
This project enables video game control using pose classification and TensorFlow Lite to create a custom controller for games.
Image UpScaler
Tensorflow implementation of a generative adversarial network for single image super-resolution
