default author photo

Congxuan Zhang

Nanchang Hangkong University

SCHOLARLY PAPERS

4

DOWNLOADS

232

TOTAL CITATIONS

0

Scholarly Papers (4)

1.

Controllable Facial Attribute Editing Via Gaussian Mixture Model Disentanglement

Posted: 30 Mar 2022
Nanchang Hangkong University, affiliation not provided to SSRN, Nanchang Hangkong University, Nanchang Hangkong University, Cardiff University, Nanchang Hangkong University and affiliation not provided to SSRN
Downloads 93 (725,554)

Abstract:

Loading...

Gaussian Mixture Model, latent Space Decoupling, Orthogonal Constraints on Manifold Spaces, Semantic Editor

2.

Certainty-Aware Disparity and Normal Estimation for High-Fidelity 3d Clothed Human Reconstruction

Number of pages: 15 Posted: 03 Mar 2025
Northwestern Polytechnic University (NPU), Northwestern Polytechnic University (NPU), Northwestern Polytechnic University (NPU), affiliation not provided to SSRN, Nanchang Hangkong University, Nanchang Hangkong University and Nanchang Hangkong University
Downloads 52 (1,041,696)

Abstract:

Loading...

Stereo matching, Human Images, Certainty-aware Refine Module, Adaptive Normal Module, 3D Human Reconstruction

3.

Plug-and-Play Super-Resolution Schedule for Pretrained  Diffusion Probability Models

Number of pages: 12 Posted: 04 Jun 2024
Bo Li, Xiao Lin, Zhi-Fen He, Kaotao Xue, Bin Liu and Congxuan Zhang
Nanchang Hangkong University, Nanchang Hangkong University, Nanchang Hangkong University, Nanchang Hangkong University, Nanchang Hangkong University and Nanchang Hangkong University
Downloads 50 (1,086,013)

Abstract:

Loading...

Plug-and-play;DDPMs;Super-resolution;Hierarchical noising schedule;Soft training strategy

4.

Diverse Style Transfer for Multiple Artists

Number of pages: 15 Posted: 28 Jun 2024
Bo Li, Wenhui Liu, Bin Liu, Zhi-Fen He, Congxuan Zhang and Yu-Kun Lai
Nanchang Hangkong University, Shenzhen University, Nanchang Hangkong University, affiliation not provided to SSRN, Nanchang Hangkong University and Cardiff University
Downloads 37 (1,222,651)

Abstract:

Loading...

style transfer, Multi-Domain, diverse synthesis, Supervised contrastive learning, Denoising diffusion probabilistic models