کنفرانس مهندسی برق ایران

صفحه اصلی / سی و یکمین کنفرانس بین المللی مهندسی برق

Transformer-Based Unsupervised Image Registration using SSIM and Homography Loss for Steady Camera and Aerial Videos

نویسندگان :

Golnoosh Abdollahinejad¹ Matin Hashemi²

1- دانشگاه صنعتی شریف 2- دانشگاه صنعتی شریف

کلمات کلیدی :

Computer Vision،Image Registration،Transformer Network،Deep Learning

چکیده :

Image registration is an essential and initial block in the pipeline of computer vision tasks and systems. It is defined as the process of transforming a moving image into a target image with minimum difference when get aligned. Unlike previous work for general-purpose datasets, aerial images suffer from mechanical shakes which leads to deformed distortion that is similar to medical volumetric image registration task. Inspired by medical approaches, we use a Transformer-based network to have a semi-unlimited receptive field Swin block, to produce a general output for each pixel named flow matrix. Flow matrix is utilized instead of regressing parameters of transformation matrix with a fixed degree of freedom that can't handle the structural difference between images. This leads to introducing of a new loss function based on the Structural Similarity Index Measure (SSIM) and embedding Homography transformation as a regularization term. The combination of a generalized-designed network and loss function based on problem definition significantly enhanced results.

لیست مقالات

لیست مقالات بایگانی شده

A Band-pass Power Divider Based on Substrate Integrated Plasmonic Waveguide

Salma Mirhadi - Shamsi Soleimani

Single-Item Fashion Recommender: Towards Cross-Domain Recommendations

Seyed Omid Mohammadi - Hossein Bodaghi - Ahmad Kalhor

Numerical Approach on Modeling of Perovskite Solar Cells Based on Coupled Ion Vacancy and Charge Carrier Dynamics

Hamed Abnavi - Daniyal Khosh Maram - Hanieh Talati Aghdam

Identifying Singular 2-D Systems Using 1-D Methods

Masoud Shafiee - Kamyar Azarakhsh

Improved Model Predictive Control for the Three-Phase Grid-Connected Split-Source Inverter

Seyed Hamid Montazeri - Jafar Milimonfared - MohammadReza Zolghadri

Improving Artificial Neural Network Performance Using Hybrid Activation Function

Morteza Taheri - Sajad Haghzad Klidbary

Automated Optic Disc Segmentation in Low-Quality Retinopathy of Prematurity Retinal Images

Abolfazl Karimiyan Abdar - Reza AghaeiZadeh Zoroofi - Naser Shoeibi - Sare Safi - Alireza Ramezani - Homayoun Nikkhah - Hamid Safi - Mohammad Reza Ansari Astaneh

Non-homogeneous interference suppression in OFDM array radars using direct data domain approach

Sima Shariatmadari

بازآرایی پویا شبکه توزیع با حضور ایستگاه شارژ خودرو الکتریکی به کمک یادگیری تقویتی

محمدامین ساعدی - صالح رازینی - محمدامین قاسمی

طبقه بندی سکته مغزی در یک سیستم دو بعدی چند فرکانسی با استفاده از امواج مایکروویو و یادگیری عمیق

محسن مهرانیان - محمدسعید ماجدی - امیررضا عطاری

بیشتر