QuickSRNet: Plain Single-Image Super-Resolution Architecture for Faster Inference on Mobile Platforms

TL;DR

QuickSRNet is a lightweight super-resolution model designed for real-time 1080p upscaling on mobile devices, achieving high accuracy with minimal latency suitable for gaming and video applications.

Contribution

The paper introduces QuickSRNet, a novel architecture that improves accuracy-to-latency trade-offs for mobile super-resolution and includes training techniques for robustness and speed.

Findings

Produces 1080p outputs via 2x upscaling in 2.2 ms on a smartphone

Outperforms existing architectures in accuracy-to-latency trade-offs

Maintains robustness to quantization

Abstract

In this work, we present QuickSRNet, an efficient super-resolution architecture for real-time applications on mobile platforms. Super-resolution clarifies, sharpens, and upscales an image to higher resolution. Applications such as gaming and video playback along with the ever-improving display capabilities of TVs, smartphones, and VR headsets are driving the need for efficient upscaling solutions. While existing deep learning-based super-resolution approaches achieve impressive results in terms of visual quality, enabling real-time DL-based super-resolution on mobile devices with compute, thermal, and power constraints is challenging. To address these challenges, we propose QuickSRNet, a simple yet effective architecture that provides better accuracy-to-latency trade-offs than existing neural architectures for single-image super resolution. We present training tricks to speed up existing residual-based super-resolution architectures while maintaining robustness to quantization. Our proposed architecture produces 1080p outputs via 2x upscaling in 2.2 ms on a modern smartphone, making it ideal for high-fps real-time applications.