Space-Time Super-Resolution using Deep Learning-based Framework


Downloads per month over past year

Sharma, M and Chaudhary, S and Lall, B (2019) Space-Time Super-Resolution using Deep Learning-based Framework. In: 7th International Conference on Pattern Recognition and Machine Intelligence, December 5-8, 2017, Kolkata. (Submitted)

Download (4Mb) | Preview


This paper introduces a novel end-to-end deep learning framework to learn the space-time super-resolution (SR) process. We propose a coupled deep convolutional auto-encoder (CDCA) which learns the non-linear mapping between convolutional features of up-sampled low- resolution (LR) video sequence patches and convolutional features of high-resolution (HR) video seQuence patches. The upsampling in LR video refers to tri-cubic interpolation both in space and time. We also propose an H.254/AVC compatible video space-time SR framework by using learned CDCA, which enables to super-resolved compressed LR video with less computational complexity. The experimental results prove that the proposed H.264/AVC compatible framework performs better than the state-of-art techniques on space-time SR in terms of quality and time complexity.

Item Type: Conference or Workshop Item (Paper)
Subjects: Electronic Systems > Digital Systems
Divisions: Electronic Systems
Depositing User: Mr. Rabin Chatterjee
Date Deposited: 04 Dec 2019 09:04
Last Modified: 04 Dec 2019 09:05

Actions (login required)

View Item View Item