Learning Dense UV Completion for 3D Human Mesh Recovery

Research output: Chapters, Conference Papers, Creative and Literary WorksRGC 32 - Refereed conference paper (with host publication)peer-review

View graph of relations

Related Research Unit(s)

Detail(s)

Original languageEnglish
Title of host publicationNeural Information Processing
Subtitle of host publication30th International Conference, ICONIP 2023, Changsha, China, November 20–23, 2023, Proceedings, Part IX
EditorsBiao Luo, Long Cheng, Zheng-Guang Wu, Hongyi Li, Chaojie Li
PublisherSpringer
Pages558-569
ISBN (Electronic)978-981-99-8138-0
ISBN (Print)978-981-99-8137-3
Publication statusPublished - 2024

Publication series

NameCommunications in Computer and Information Science
Volume1963
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Title30th International Conference on Neural Information Processing (ICONIP 2023)
PlaceChina
CityChangsha
Period20 - 23 November 2023

Abstract

Human mesh reconstruction from a single image is a challenging task due to the occlusion caused by self, objects, or other humans. Existing methods either fail to separate human features accurately or lack proper supervision for feature completion. In this paper, we propose Dense Inpainting Human Mesh Recovery (DIMR), a two-stage method that leverages dense correspondence maps to handle occlusion. Our method utilizes a dense correspondence map to separate visible human features and completes human features on a structured UV space with an attention-based feature completion module. We also design a feature inpainting training procedure that guides the network to learn from unoccluded features. We evaluate our method on several datasets and demonstrate its superior performance under heavily occluded scenarios compared to other methods. Extensive experiments show that our method obviously outperforms prior methods on heavily occluded images and achieves comparable results on the standard benchmarks. Moreover, our method is comparable with previous methods on no heavily occluded images. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

Research Area(s)

  • Inpainting, IUV, Occlusion, Skinned Multi-Person Linear Model (SMPL)

Citation Format(s)

Learning Dense UV Completion for 3D Human Mesh Recovery. / Sun, Qingping; Wang, Yanjun; Wang, Zhenni et al.
Neural Information Processing: 30th International Conference, ICONIP 2023, Changsha, China, November 20–23, 2023, Proceedings, Part IX. ed. / Biao Luo; Long Cheng; Zheng-Guang Wu; Hongyi Li; Chaojie Li. Springer, 2024. p. 558-569 (Communications in Computer and Information Science; Vol. 1963).

Research output: Chapters, Conference Papers, Creative and Literary WorksRGC 32 - Refereed conference paper (with host publication)peer-review