Long Feng

Mr. Long Feng


Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
29 May 2024 in Journal of Imaging
Verified
Journal Article
Open access
PDF file
YRYong Ren
Yong Ren
MZMingquan Zhou
Mingquan Zhou
YWYifan Wang
Yifan Wang
LFLong Feng
Long Feng
QZQiuquan Zhu
Qiuquan Zhu
GGGuohua Geng
Guohua Geng
Accurate and robust 3D human modeling from a single image presents significant challenges. Existing methods have shown potential, but they often fail to generate reconstructions that match the level of detail in the input image. These methods particularly struggle with loose clothing. They typically employ parameterized human models to constrain the reconstruction process, ensuring the results do not deviate too far from the model and produce anomalies. However, this also limits the recovery of loose clothing. To address this issue, we propose an end-to-end method called IHRPN for reconstructing clothed humans from a single 2D human image. This method includes a feature extraction module for semantic extraction of image features. We propose an image semantic feature extraction aimed at achieving pixel model space consistency and enhancing the robustness of loose clothing. We extract features from the input image to infer and recover the SMPL-X mesh, and then combine it with a normal map to guide the implicit function to reconstruct the complete clothed human. Unlike traditional methods, we use local features for implicit surface regression. Our experimental results show that our IHRPN method performs excellently on the CAPE and AGORA datasets, achieving good performance, and the reconstruction of loose clothing is noticeably more accurate and robust.

12 February 2024
Verified
Book Chapter
Open access
PDF file
YZYanting Zhao
Yanting Zhao
GGGuohua Geng
Guohua Geng
LFLong Feng
Long Feng
HWHaodong Wu
Haodong Wu
QLQihang Li
Qihang Li
To apply makeup style of half-body reference images with complex makeup styles to the source image, we propose a model that combines object detection and makeup transfer. This model uses MTCNN to detect the target of Qinqiang Opera makeup image, and obtains the facial image with only makeup, so as to prevent the effect of makeup migration from being blocked by hair ornaments or hairstyles. And use the facial image obtained from object detection that only contains makeup as the reference image for the makeup transfer section, and use the CPM model to obtain the final result of makeup transfer in the makeup transfer section. In addition, we also collected the Qinqiang Opera character makeup dataset QinqMakeup. The Qinqiang Opera makeup migration method can not only enhance the entertainment of Qinqiang Opera, but also attract young groups and expand its audience. The ancient Qinqiang Opera art will be integrated with modern digital technology to achieve the inheritance and development of Qinqiang Opera, a Intangible cultural heritage.