Deconvolution of Image Sequences with a Learning FFT-based Approach

Iaroslav Koshelev, Andrey Somov, Stamatios Lefkimmiatis, Antonio Rodriguez-Sanchez

Результат исследований: Глава в книге, отчете, сборнике статейМатериалы для конференциирецензирование

1 Цитирования (Scopus)

Аннотация

In this work, we address the problem of fast and accurate non-blind motion deblurring of a sequence of frames. We propose an approach based on single image non-blind deconvolution methods by extending them from two spatial dimensions onto a 2D+ time space. The resulted algorithms utilize recent advances of deep learning and heavily rely on learning the spatial and time correlations of frames from data which allows to perform the deblurring in an end-to-end fashion without the manual tuning of parameters. Developed algorithms were trained and tested on blurred video sequences, and as a result an improvement in restoration quality by around 1dB in terms of PSNR was achieved, comparing to a single-frame deconvolution. The proposed algorithms perform video restoration in a close to real-time speed, having potential for real applications in both portable consumer electronics and machine vision systems.

Язык оригиналаАнглийский
Название основной публикации2020 IEEE 29th International Symposium on Industrial Electronics, ISIE 2020 - Proceedings
ИздательInstitute of Electrical and Electronics Engineers Inc.
Страницы1381-1386
Число страниц6
ISBN (электронное издание)9781728156354
DOI
СостояниеОпубликовано - июн. 2020
Опубликовано для внешнего пользованияДа
Событие29th IEEE International Symposium on Industrial Electronics, ISIE 2020 - Delft, Голландия
Продолжительность: 17 июн. 202019 июн. 2020

Серия публикаций

НазваниеIEEE International Symposium on Industrial Electronics
Том2020-June

Конференция

Конференция29th IEEE International Symposium on Industrial Electronics, ISIE 2020
Страна/TерриторияГолландия
ГородDelft
Период17/06/2019/06/20

Fingerprint

Подробные сведения о темах исследования «Deconvolution of Image Sequences with a Learning FFT-based Approach». Вместе они формируют уникальный семантический отпечаток (fingerprint).

Цитировать