Previous Next

新闻中心


段延松作为共同通讯在 《Information Processing & Management》上发表文章: Haar-wavelet based texture inpainting for human pose transfer

发布时间:2024-03-12 17:38 作者: 来源: 阅读:

段延松作为共同通讯在 《Information Processing & Management》上发表文章:

Haar-wavelet based texture inpainting for human pose transfer


欢迎引用

Cite:

Wu Haoran, He Fazhi, Duan Yansong, et al.Haar-wavelet based texture inpainting for human pose transfer[J].INFORMATION PROCESSING & MANAGEMENT,2024,61(3).



Haar-wavelet based texture inpainting for human pose transfer

https://doi.org/10.1016/j.ipm.2023.103612Get rights and content

Highlights

    • A novel wavelet transformation-based encoder-decoder structure to mitigate the ill-posed problems.

    • A novel architecture HBM that utilizes the high-frequency information from the input source image.

    • Our method can preserve high-frequency contents in the human image for challenging generation task.

    • Our work is a fundamental building block in various computer vision applications and engineering tasks.

Abstract

Human Image Generation is an important information processing technique. Producing realistic-looking texture is crucial for Generative Adversarial Networks (GAN) based person image generation. Existing methods follow a “downscale-upscale” strategy that source images are usually downscaled to extract features and saved the cost of storage. Meanwhile, the upscaling process is applied to recover the details based on these features. The loss of high-frequency components during the downscaling process, however, is in accordance with the Nyquist-Shannon sampling theorem, which creates an ill-posed difficulty during the upscaling process. In this paper, we design a Haar-wavelet based texture inpainting network (HWTIN) to mitigate the ill-posed problem in pose transfer task. In the downscaling process, to divide the source image into high-frequency and low-frequency contents, we construct a Haar-based Wavelet Module (HBM). In this way, We can preserve these high-frequency information in the generation process. We also design an inverse HBM (IHBM) to utilize these high-frequency information in the upscaling process. Extensive results on mainstream datasets demonstrate that HWTIN outperforms state-of-the-art (SOTA) methods quantitatively.