Real-ESRGAN-x4plus is a high-performance image super-resolution model based on the ESRGAN (Enhanced Super-Resolution Generative Adversarial Networks) architecture. It can upscale low-resolution images by 4 times while significantly enhancing details and visual quality. The model uses an improved GAN structure to optimize detail restoration and noise reduction, making it suitable for photo restoration, video enhancement, and other applications. Real-ESRGAN-x4plus supports various hardware accelerations for efficient inference on mobile and edge computing platforms.

Source model

• Input shape: 1x3x128x128
• Number of parameters: 15.92M
• Model size: 67.76M
• Output shape: 1x3x512x512

The source model can be found here

Click Model Conversion Reference in the Performance Reference panel on the right to view the model conversion steps.

The model performance benchmarks and inference example code provided on Model Farm are all based on the APLUX AidLite SDK

SDK installation

For details, please refer to the AidLite Developer Documentation

• Install AidLite SDK

# install aidlite sdk c++ api
sudo aid-pkg -i aidlite-sdk

# install aidlite sdk python api
python3 -m pip install pyaidlite -i https://mirrors.aidlux.com --trusted-host mirrors.aidlux.com

• Verify AidLite SDK

# aidlite sdk c++ check
python3 -c "import aidlite; print(aidlite.get_library_version())"

# aidlite sdk python check
python3 -c "import aidlite; print(aidlite.get_py_library_version())"

Inference example

• Click Model & Test Code to download model files and inference codes. The file structure showed below:

/model_farm_{model_name}_aidlite
    
    |__ models # folder where model files are stored

    |__ python # aidlite python model inference example

    |__ cpp # aidlite cpp model inference example

    |__ README.md