Learning Long-Term Style-Preserving Blind Video Temporal Consistency

03/12/2021
by   Hugo Thimonier, et al.
27

When trying to independently apply image-trained algorithms to successive frames in videos, noxious flickering tends to appear. State-of-the-art post-processing techniques that aim at fostering temporal consistency, generate other temporal artifacts and visually alter the style of videos. We propose a postprocessing model, agnostic to the transformation applied to videos (e.g. style transfer, image manipulation using GANs, etc.), in the form of a recurrent neural network. Our model is trained using a Ping Pong procedure and its corresponding loss, recently introduced for GAN video generation, as well as a novel style preserving perceptual loss. The former improves long-term temporal consistency learning, while the latter fosters style preservation. We evaluate our model on the DAVIS and videvo.net datasets and show that our approach offers state-of-the-art results concerning flicker removal, and better keeps the overall style of the videos than previous approaches.

READ FULL TEXT

page 1

page 5

page 6

research
07/03/2018

ReCoNet: Real-time Coherent Video Style Transfer Network

Image style transfer models based on convolutional neural networks usual...
research
11/27/2019

LucidDream: Controlled Temporally-Consistent DeepDream on Videos

In this work, we aim to propose a set of techniques to improve the contr...
research
06/08/2022

Learning Task Agnostic Temporal Consistency Correction

Due to the scarcity of video processing methodologies, image processing ...
research
08/01/2018

Learning Blind Video Temporal Consistency

Applying image processing algorithms independently to each frame of a vi...
research
07/11/2022

CCPL: Contrastive Coherence Preserving Loss for Versatile Style Transfer

In this paper, we aim to devise a universally versatile style transfer m...
research
05/22/2019

LapTool-Net: A Contextual Detector of Surgical Tools in Laparoscopic Videos Based on Recurrent Convolutional Neural Networks

We propose a new multilabel classifier, called LapTool-Net to detect the...
research
01/23/2021

BSUV-Net 2.0: Spatio-Temporal Data Augmentations for Video-AgnosticSupervised Background Subtraction

Background subtraction (BGS) is a fundamental video processing task whic...

Please sign up or login with your details

Forgot password? Click here to reset