Collapsible Linear Blocks for Super-Efficient Super Resolution

03/17/2021
by   Kartikeya Bhardwaj, et al.
5

With the advent of smart devices that support 4K and 8K resolution, Single Image Super Resolution (SISR) has become an important computer vision problem. However, most super resolution deep networks are computationally very expensive. In this paper, we propose SESR, a new class of Super-Efficient Super Resolution networks that significantly improve image quality and reduce computational complexity. Detailed experiments across six benchmark datasets demonstrate that SESR achieves similar or better image quality than state-of-the-art models while requiring 2x to 330x fewer Multiply-Accumulate (MAC) operations. As a result, SESR can be used on constrained hardware to perform x2 (1080p to 4K) and x4 SISR (1080p to 8K). Towards this, we simulate hardware performance numbers for a commercial mobile Neural Processing Unit (NPU) for 1080p to 4K (x2) and 1080p to 8K (x4) SISR. Our results highlight the challenges faced by super resolution on AI accelerators and demonstrate that SESR is significantly faster than existing models. Overall, SESR establishes a new Pareto frontier on the quality (PSNR)-computation relationship for the super resolution task.

READ FULL TEXT

page 3

page 7

page 8

page 11

page 12

research
01/31/2018

SESR: Single Image Super Resolution with Recursive Squeeze and Excitation Networks

Single image super resolution is a very important computer vision task, ...
research
12/29/2021

Super-Efficient Super Resolution for Fast Adversarial Defense at the Edge

Autonomous systems are highly vulnerable to a variety of adversarial att...
research
08/30/2019

Robust Online Video Super-Resolution Using an Efficient Alternating Projections Scheme

Video super-resolution reconstruction (SRR) algorithms attempt to recons...
research
09/07/2022

Generative Adversarial Super-Resolution at the Edge with Knowledge Distillation

Single-Image Super-Resolution can support robotic tasks in environments ...
research
02/27/2023

LSR: A Light-Weight Super-Resolution Method

A light-weight super-resolution (LSR) method from a single image targeti...
research
10/13/2019

ERNet Family: Hardware-Oriented CNN Models for Computational Imaging Using Block-Based Inference

Convolutional neural networks (CNNs) demand huge DRAM bandwidth for comp...
research
03/02/2020

Gated Fusion Network for Degraded Image Super Resolution

Single image super resolution aims to enhance image quality with respect...

Please sign up or login with your details

Forgot password? Click here to reset