New Evaluation Method of Animation Image Transfer

DOI

10.2991/assehr.k.220504.069How to use a DOI?

Keywords

Machine Learning; Cycle-Gan; Python; evaluation

Abstract

Today, many animation industries need to use landscape migration technology. However, the evaluation metrics after landscape transfer are not particularly accurate, which leads to some failed landscape transfer images being used in the final animation pages. People are reluctant to see that phenomenon. Therefore, it is necessary to find a judgment standard suitable for landscape indicators, and select qualified pictures after landscape migration. Therefore, we will use Cycle-Gan as a method to convert graphics by using the Python programming language and then adjust the weights of psnr and ssin as the latest evaluation indicators in order to mitigate the uncorrected transferred images that appear at the final result. We finally use a new metric as a judging criterion to get high-quality pictures after landscape migration.

Copyright

© 2022 The Authors. Published by Atlantis Press SARL.

Open Access

This is an open access article distributed under the CC BY-NC 4.0 license.

Download article (PDF)

TY  - CONF
AU  - Dixin Li
PY  - 2022
DA  - 2022/06/01
TI  - New Evaluation Method of Animation Image Transfer
BT  - Proceedings of the 2022 8th International Conference on Humanities and Social Science Research (ICHSSR 2022)
PB  - Atlantis Press
SP  - 371
EP  - 376
SN  - 2352-5398
UR  - https://doi.org/10.2991/assehr.k.220504.069
DO  - 10.2991/assehr.k.220504.069
ID  - Li2022
ER  -