Self-supervised recurrent depth estimation with attention mechanisms

Author:

Makarov Ilya123,Bakhanova Maria1,Nikolenko Sergey45,Gerasimova Olga1

Affiliation:

1. HSE University, Moscow, Russia

2. Artificial Intelligence Research Institute (AIRI), Moscow, Russia

3. Big Data Research Center, National University of Science and Technology MISIS, Moscow, Russia

4. Steklov Institute of Mathematics at St. Petersburg, St. Petersburg, Russia

5. St. Petersburg State University, St. Petersburg, Russia

Abstract

Depth estimation has been an essential task for many computer vision applications, especially in autonomous driving, where safety is paramount. Depth can be estimated not only with traditional supervised learning but also via a self-supervised approach that relies on camera motion and does not require ground truth depth maps. Recently, major improvements have been introduced to make self-supervised depth prediction more precise. However, most existing approaches still focus on single-frame depth estimation, even in the self-supervised setting. Since most methods can operate with frame sequences, we believe that the quality of current models can be significantly improved with the help of information about previous frames. In this work, we study different ways of integrating recurrent blocks and attention mechanisms into a common self-supervised depth estimation pipeline. We propose a set of modifications that utilize temporal information from previous frames and provide new neural network architectures for monocular depth estimation in a self-supervised manner. Our experiments on the KITTI dataset show that proposed modifications can be an effective tool for exploiting temporal information in a depth prediction pipeline.

Funder

HSE University Basic Research Program

Publisher

PeerJ

Subject

General Computer Science

Reference65 articles.

1. Delving deeper into convolutional networks for learning video representations;Ballas,2015

2. Estimating depth from monocular images as classification using deep fully convolutional residual networks;Cao,2016

3. Depth prediction without the sensors: leveraging structure for unsupervised learning from monocular videos;Casser,2018

4. Learning phrase representations using RNN encoder-decoder for statistical machine translation;Cho,2014

5. Predicting depth, surface normals and semantic labels with a common multi-scale convolutional architecture;Eigen,2014

Cited by 27 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Polarimetric Imaging for Robot Perception: A Review;Sensors;2024-07-09

2. mini-Unet GAN: Optimized GAN for Monocular Depth Estimation;2024 6th International Conference on Pattern Analysis and Intelligent Systems (PAIS);2024-04-24

3. Inpainting Semantic and Depth Features to Improve Visual Place Recognition in the Wild;IEEE Access;2024

4. Attention U-Net Oriented Towards 3D Depth Estimation;Lecture Notes in Networks and Systems;2024

5. Application of Multimodal Machine Learning for Image Recommendation Systems;Communications in Computer and Information Science;2024

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3