A comprehensive survey on hardware-aware neural architecture search

H Benmeziane, KE Maghraoui, H Ouarnoughi… - arxiv preprint arxiv …, 2021 - arxiv.org
Neural Architecture Search (NAS) methods have been growing in popularity. These
techniques have been fundamental to automate and speed up the time consuming and error …

Weight-sharing neural architecture search: A battle to shrink the optimization gap

L **e, X Chen, K Bi, L Wei, Y Xu, L Wang… - ACM Computing …, 2021 - dl.acm.org
Neural architecture search (NAS) has attracted increasing attention. In recent years,
individual search methods have been replaced by weight-sharing search methods for higher …

Edge-oriented convolution block for real-time super resolution on mobile devices

X Zhang, H Zeng, L Zhang - Proceedings of the 29th ACM International …, 2021 - dl.acm.org
Efficient and light-weight super resolution (SR) is highly demanded in practical applications.
However, most of the existing studies focusing on reducing the number of model parameters …

Fourier space losses for efficient perceptual image super-resolution

D Fuoli, L Van Gool, R Timofte - Proceedings of the IEEE …, 2021 - openaccess.thecvf.com
Many super-resolution (SR) models are optimized for high performance only and therefore
lack efficiency due to large model complexity. As large models are often not practical in real …

Contextual transformation network for lightweight remote-sensing image super-resolution

S Wang, T Zhou, Y Lu, H Di - IEEE Transactions on Geoscience …, 2021 - ieeexplore.ieee.org
Current super-resolution networks typically reduce network parameters and multiadds
operations by designing lightweight structures, but lightening the convolution layer is often …

Practical single-image super-resolution using look-up table

Y Jo, SJ Kim - Proceedings of the IEEE/CVF Conference on …, 2021 - openaccess.thecvf.com
A number of super-resolution (SR) algorithms from in terpolation to deep neural networks
(DNN) have emerged to restore or create missing details of the input low-resolution image …

Hitchhiker's guide to super-resolution: Introduction and recent advances

BB Moser, F Raue, S Frolov, S Palacio… - … on Pattern Analysis …, 2023 - ieeexplore.ieee.org
With the advent of Deep Learning (DL), Super-Resolution (SR) has also become a thriving
research area. However, despite promising results, the field still faces challenges that …

Nas-bench-asr: Reproducible neural architecture search for speech recognition

A Mehrotra, AGCP Ramos, S Bhattacharya… - International …, 2021 - openreview.net
Powered by innovations in novel architecture design, noise tolerance techniques and
increasing model capacity, Automatic Speech Recognition (ASR) has made giant strides in …

Repsr: Training efficient vgg-style super-resolution networks with structural re-parameterization and batch normalization

X Wang, C Dong, Y Shan - Proceedings of the 30th ACM International …, 2022 - dl.acm.org
This paper explores training efficient VGG-style super-resolution (SR) networks with the
structural re-parameterization technique. The general pipeline of re-parameterization is to …

Learning series-parallel lookup tables for efficient image super-resolution

C Ma, J Zhang, J Zhou, J Lu - European Conference on Computer Vision, 2022 - Springer
Lookup table (LUT) has shown its efficacy in low-level vision tasks due to the valuable
characteristics of low computational cost and hardware independence. However, recent …