Saved in:
Bibliographic Details
Main Authors: Singh, Ritambhara, Jain, Abhishek, Perona, Pietro, Agarwal, Shivani, Yang, Junfeng
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.05398
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911773521608704
author Singh, Ritambhara
Jain, Abhishek
Perona, Pietro
Agarwal, Shivani
Yang, Junfeng
author_facet Singh, Ritambhara
Jain, Abhishek
Perona, Pietro
Agarwal, Shivani
Yang, Junfeng
contents High-resolution semantic segmentation requires substantial computational resources. Traditional approaches in the field typically downscale the input images before processing and then upscale the low-resolution outputs back to their original dimensions. While this strategy effectively identifies broad regions, it often misses finer details. In this study, we demonstrate that a streamlined model capable of directly producing high-resolution segmentations can match the performance of more complex systems that generate lower-resolution results. By simplifying the network architecture, we enable the processing of images at their native resolution. Our approach leverages a bottom-up information propagation technique across various scales, which we have empirically shown to enhance segmentation accuracy. We have rigorously tested our method using leading-edge semantic segmentation datasets. Specifically, for the Cityscapes dataset, we further boost accuracy by applying the Noisy Student Training technique.
format Preprint
id arxiv_https___arxiv_org_abs_2402_05398
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle On the Effect of Image Resolution on Semantic Segmentation
Singh, Ritambhara
Jain, Abhishek
Perona, Pietro
Agarwal, Shivani
Yang, Junfeng
Computer Vision and Pattern Recognition
High-resolution semantic segmentation requires substantial computational resources. Traditional approaches in the field typically downscale the input images before processing and then upscale the low-resolution outputs back to their original dimensions. While this strategy effectively identifies broad regions, it often misses finer details. In this study, we demonstrate that a streamlined model capable of directly producing high-resolution segmentations can match the performance of more complex systems that generate lower-resolution results. By simplifying the network architecture, we enable the processing of images at their native resolution. Our approach leverages a bottom-up information propagation technique across various scales, which we have empirically shown to enhance segmentation accuracy. We have rigorously tested our method using leading-edge semantic segmentation datasets. Specifically, for the Cityscapes dataset, we further boost accuracy by applying the Noisy Student Training technique.
title On the Effect of Image Resolution on Semantic Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2402.05398