Temporal-Aware Chinese Ink-Wash Style Transfer for 3D Scenes

Chinese ink-wash stylization method for 3D scenes based on neural style transfer lack temporal consistency and produce flickering artifacts. We propose a Chinese ink-wash style transfer method that specialize in 3D scene stylization with temporal aware ability. Our method transfers the artistic style of ink-wash paintings to render frames in 3D scenes, and by incorporating additional losses that measure temporal consistency in terms of stylized outputs, image features and depth maps, our method ensure necessary image coherency between consecutive frames from multiple aspects. Comparing to state-of-the-art studies, our method achieved more temporally stable results both in qualitative and quantitative experiments.

Authors

Institutions

Publication Details

Published
2025-05-23
DOI
https://doi.org/10.1109/cvidl65390.2025.11085885
Primary Topic
Generative Adversarial Networks and Image Synthesis
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Temporal-Aware Chinese Ink-Wash Style Transfer for 3D Scenes

Wenzhe Zhu, Peng Yu, Qing Zhu
Generative Adversarial Networks and Image Synthesis
article

Temporal-Aware Chinese Ink-Wash Style Transfer for 3D Scenes

Wenzhe Zhu, Peng Yu, Qing Zhu
article en

Abstract

Chinese ink-wash stylization method for 3D scenes based on neural style transfer lack temporal consistency and produce flickering artifacts. We propose a Chinese ink-wash style transfer method that specialize in 3D scene stylization with temporal aware ability. Our method transfers the artistic style of ink-wash paintings to render frames in 3D scenes, and by incorporating additional losses that measure temporal consistency in terms of stylized outputs, image features and depth maps, our method ensure necessary image coherency between consecutive frames from multiple aspects. Comparing to state-of-the-art studies, our method achieved more temporally stable results both in qualitative and quantitative experiments.

Beijing University of Technology (CN)
Natural Science Foundation of Beijing Municipality
Openalex Percentile: Top 84%
Generative Adversarial Networks and Image Synthesis
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.