Generative Adversarial Super-Resolution at the Edge with Knowledge Distillation
This work addresses the need for reliable visual streams in mobile robotic applications, such as mission monitoring and teleoperation, by providing a fast and lightweight super-resolution solution, though it is incremental as it builds on existing SRGAN architecture with optimizations.
The authors tackled the problem of real-time single-image super-resolution for robotic applications by proposing EdgeSRGAN, an efficient generative adversarial network model optimized for CPU and Edge TPU devices, achieving up to 200 fps inference while preserving image quality comparable to heavier state-of-the-art models.
Single-Image Super-Resolution can support robotic tasks in environments where a reliable visual stream is required to monitor the mission, handle teleoperation or study relevant visual details. In this work, we propose an efficient Generative Adversarial Network model for real-time Super-Resolution, called EdgeSRGAN (code available at https://github.com/PIC4SeR/EdgeSRGAN). We adopt a tailored architecture of the original SRGAN and model quantization to boost the execution on CPU and Edge TPU devices, achieving up to 200 fps inference. We further optimize our model by distilling its knowledge to a smaller version of the network and obtain remarkable improvements compared to the standard training approach. Our experiments show that our fast and lightweight model preserves considerably satisfying image quality compared to heavier state-of-the-art models. Finally, we conduct experiments on image transmission with bandwidth degradation to highlight the advantages of the proposed system for mobile robotic applications.