A Fast 4K Video Frame Interpolation Using a Hybrid Task-Based Convolutional Neural Network

Visual quality and algorithm efficiency are two main interests in video frame interpolation. We propose a hybrid task-based convolutional neural network for fast and accurate frame interpolation of 4K videos. The proposed method synthesizes low-resolution frames, then reconstructs high-resolution fr...

Full description

Saved in:
Bibliographic Details
Published in:Symmetry (Basel) 2019-05, Vol.11 (5), p.619
Main Authors: Ahn, Ha-Eun, Jeong, Jinwoo, Kim, Je Woo
Format: Article
Language:English
Subjects:
Accuracy
Algorithms
Artificial neural networks
edge loss
frame interpolation
Frames (data processing)
high-resolution image processing
hybrid network
Interpolation
Methods
Neural networks
super-resolution
Synthesis
Video
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Visual quality and algorithm efficiency are two main interests in video frame interpolation. We propose a hybrid task-based convolutional neural network for fast and accurate frame interpolation of 4K videos. The proposed method synthesizes low-resolution frames, then reconstructs high-resolution frames in a coarse-to-fine fashion. We also propose edge loss, to preserve high-frequency information and make the synthesized frames look sharper. Experimental results show that the proposed method achieves state-of-the-art performance and performs 2.69x faster than the existing methods that are operable for 4K videos, while maintaining comparable visual and quantitative quality.
ISSN:2073-8994
2073-8994
DOI:10.3390/sym11050619