Mobile-friendly Image de-noising: Hardware Conscious Optimization for Edge Application

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Miriyala, Srinivas, Vajrala, Sowmya, Kumar, Hitesh, Kodavanti, Sravanth, Rajendiran, Vikram
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915736740429824
author Miriyala, Srinivas
Vajrala, Sowmya
Kumar, Hitesh
Kodavanti, Sravanth
Rajendiran, Vikram
author_facet Miriyala, Srinivas
Vajrala, Sowmya
Kumar, Hitesh
Kodavanti, Sravanth
Rajendiran, Vikram
contents Image enhancement is a critical task in computer vision and photography that is often entangled with noise. This renders the traditional Image Signal Processing (ISP) ineffective compared to the advances in deep learning. However, the success of such methods is increasingly associated with the ease of their deployment on edge devices, such as smartphones. This work presents a novel mobile-friendly network for image de-noising obtained with Entropy-Regularized differentiable Neural Architecture Search (NAS) on a hardware-aware search space for a U-Net architecture, which is first-of-its-kind. The designed model has 12% less parameters, with ~2-fold improvement in ondevice latency and 1.5-fold improvement in the memory footprint for a 0.7% drop in PSNR, when deployed and profiled on Samsung Galaxy S24 Ultra. Compared to the SOTA Swin-Transformer for Image Restoration, the proposed network had competitive accuracy with ~18-fold reduction in GMACs. Further, the network was tested successfully for Gaussian de-noising with 3 intensities on 4 benchmarks and real-world de-noising on 1 benchmark demonstrating its generalization ability.
format Preprint
id arxiv_https___arxiv_org_abs_2601_11684
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Mobile-friendly Image de-noising: Hardware Conscious Optimization for Edge Application
Miriyala, Srinivas
Vajrala, Sowmya
Kumar, Hitesh
Kodavanti, Sravanth
Rajendiran, Vikram
Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
Image enhancement is a critical task in computer vision and photography that is often entangled with noise. This renders the traditional Image Signal Processing (ISP) ineffective compared to the advances in deep learning. However, the success of such methods is increasingly associated with the ease of their deployment on edge devices, such as smartphones. This work presents a novel mobile-friendly network for image de-noising obtained with Entropy-Regularized differentiable Neural Architecture Search (NAS) on a hardware-aware search space for a U-Net architecture, which is first-of-its-kind. The designed model has 12% less parameters, with ~2-fold improvement in ondevice latency and 1.5-fold improvement in the memory footprint for a 0.7% drop in PSNR, when deployed and profiled on Samsung Galaxy S24 Ultra. Compared to the SOTA Swin-Transformer for Image Restoration, the proposed network had competitive accuracy with ~18-fold reduction in GMACs. Further, the network was tested successfully for Gaussian de-noising with 3 intensities on 4 benchmarks and real-world de-noising on 1 benchmark demonstrating its generalization ability.
title Mobile-friendly Image de-noising: Hardware Conscious Optimization for Edge Application
topic Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2601.11684