0% Complete
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 Abdollahinejad
1
Matin Hashemi
2
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
Improving CycleGAN-VC2 Voice Conversion by Learning MCD-Based Evaluation and Optimization
Majid Behdad - Davood Gharavian
Probabilistic Dynamic Economic Dispatch in Presence of Wind Farms
Homayoun Berahmandpour - Shahram Montasar Kuhsari - Hassan Rastegar
Design of Dual-Band Triangular Microstrip Antenna Using Fractal Structure for Wi-Max and Wi-Fi Applications
Arian Mianji - Mohammad Bemani - Saeid Nikmehr - Ahmad Atashpaz Gargari
Underwater Image Quality Assessment via Color and Contrast Analysis
Meysam Ghalyani - Maryam Karimi
انتخاب سبد سهام بهینه در بورس تهران با استفاده از تقریب تصادفی انحراف همزمان
زینب گدازگر
تحلیل عدم تعادل جریان سه فاز شبکه فشارضعیف توزیع در پی قطع هادی نول متصل به ترانسفورماتور با استفاده از مولفههای متقارن
احمد صالحی دوبخشری
Age of Information Optimization for Multi-hop VLC/RF IoT Sensor Networks
Hossein Khodi - Paeiz Azmi - Nader Mokari - Mohammadreza Javan - Hamid Saeedi - Murat Uysal
On the Impact of Probabilistic Shaping on Fiber Nonlinearities
Ahmad Tanha - Hami Rabbani - Lotfollah Beygi
Noninvasive Diagnosis of the Type of Breast Tumor through Artificial Neural Networks
Pooya Tahmasebi - Maryam Mehdizadeh Dastjerdi - Ali Fallah - Saeid Rashidi
A New Unsupervised Feature Learning Method for Object Recognition using Prior-Knowledge Data
Ashkan Farrokhi - Hadi Seyedarabi
ثمین همایش، سامانه مدیریت کنفرانس ها و جشنواره ها - نگارش 40.4.2