Improving Video Colorization by Test-Time Tuning

Yaping Zhao, Haitian Zheng, Jiebo Luo, Edmund Y. Lam
6/25/2023
eess.IVcs.CV

Official code: IndigoPurple/T3 (2 stars, MIT, last commit 2025-01-04).

Abstract

With the advancements in deep learning, video colorization by propagating color information from a colorized reference frame to a monochrome video sequence has been well explored. However, the existing approaches often suffer from overfitting the training dataset and sequentially lead to suboptimal performance on colorizing testing samples. To address this issue, we propose an effective method, which aims to enhance video colorization through test-time tuning. By exploiting the reference to construct additional training samples during testing, our approach achieves a performance boost of 1~3 dB in PSNR on average compared to the baseline. Code is available at: https://github.com/IndigoPurple/T3

View on arXivView PDF

Code Implementations(1)

IndigoPurple/T3Official100%
MIT

Python (PyTorch) code and model of "Improving Video Colorization by Test-Time Tuning"

20Shell, PythonJun 23, 20231 years agoMIT

Similar Papers

Video Colorization with Pre-trained Text-to-Image Diffusion Models

86% match2023

LatentColorization: Latent Diffusion-Based Speaker Video Colorization

85% match2024

Temporal Consistent Automatic Video Colorization via Semantic Correspondence

84% match1 repo92023

Prompt-based Consistent Video Colorization

84% match2025

L-C4: Language-Based Video Colorization for Creative and Consistent Color

83% match2024

VanGogh: A Unified Multimodal Diffusion-based Framework for Video Colorization

83% match2025

Cite this paper

@article{zhao2023improving,
  title  = {Improving Video Colorization by Test-Time Tuning},
  author = {Yaping Zhao and Haitian Zheng and Jiebo Luo and Edmund Y. Lam},
  year   = {2023},
  eprint = {2307.11757},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2307.11757v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2307.11757.svg)](https://codeofpaper.com/paper/2307.11757)

Discussion