|
I am a third-year Eng.D. student of Huazhong University of Science and Technology, School of CS. I got my bachelor degree from HUST in 2023. Under the supervised of Prof. Xinggang Wang, Prof. Wenyu Liu in the School of EIC, I am enjoying my Eng.D. career in collaboration with Huawei Inc. My research interests mainly focus on 3D Vision/ Neural Rendering. |
|
|
|
My research interests are efficient neural rendering technology, including: 4D Reconstruction, 3D Generation etc. * Equal contribution. |
|
 |
Guanjun Wu*, Jiemin Fang*, Chen Yang*, Sikuang Li, Taoran Yi, Jia Lu, Zanwei Zhou, Jiazhong Cen, Lingxi Xie, Xiaopeng Zhang, Wei Wei, Wenyu Liu, Xinggang Wang, Qi Tian, Arxiv, 2025 [ Paper] [ Page] [ Code][ Talk] We propose UniLat3D, a novel one-stage 3D generation framework, which unifies geometry and appearance in a compact latent representation, achieving superior performance than common two-stage 3D generation models. |
 |
Guanjun Wu*, Taoran Yi*, Jiemin Fang, Lingxi Xie, Xiaopeng Zhang, Wei Wei, Wenyu Liu, Qi Tian, Xinggang Wang CVPR, 2024 [ Paper] [ Page] [ Code] [ Talk] Ranked Top5 in CVPR paper influence, 3000+ GitHub stars. In 4D-GS, a novel explicit representation containing both 3D Gaussians and 4D neural voxels is proposed. A decomposed neural voxel encoding algorithm inspired by HexPlane is proposed to efficiently build Gaussian features from 4D neural voxels and then a lightweight MLP is applied to predict Gaussian deformations at novel timestamps. Our 4D-GS method achieves real-time rendering under high resolutions, 82 FPS at an 800 X 800 resolution on an RTX 3090 GPU while maintaining comparable or better quality than previous state-of-the-art methods. |
 |
Guanjun Wu*, Taoran Yi*, Jiemin Fang, Wenyu Liu, Xinggang Wang 3DV, 2024 [ Paper] [ Page] [ Code] [ Data] we propose a dynamic HDR NeRF framework, named as HDR-HexPlane, which can learn 3D scenes from dynamic 2D images captured with various exposures.With the proposed model, high-quality novel-view images at any time point can be rendered with any desired exposure. We further construct a dataset containing multiple dynamic scenes captured with diverse exposures for evaluation. |
 |
Shuai Zhang*, Guanjun Wu*, Zhoufeng Xie, Xinggang Wang, Bin Feng, Wenyu Liu ACMMM, 2025 (Oral) [ Paper] [ Code] we propose a novel representation that can reconstruct accurate meshes from sparse image input, named Dynamic 2D Gaussians (D-2DGS). We adopt 2D Gaussians for basic geometry representation and use sparse-controlled points to capture 2D Gaussian's deformation. By extracting the object mask from the rendered high-quality image and masking the rendered depth map, a high-quality dynamic mesh sequence of the object can be extracted. |
|
|
|
 |
Taoran Yi, Jiemin Fang, Zanwei Zhou, Junjie Wang, Guanjun Wu, Lingxi Xie, Xiaopeng Zhang, Wenyu Liu, Xinggang Wang, Qi Tian ArXiv, 2024 [ Paper] [ Page] [ Code] Aiming at highly enhancing the generation quality, we propose a novel framework named GaussianDreamerPro. The main idea is to bind Gaussians to reasonable geometry, which evolves over the whole generation process. Along different stages of our framework, both the geometry and appearance can be enriched progressively. The final output asset is constructed with 3D Gaussians bound to mesh, which shows significantly enhanced details and quality compared with previous methods. Notably, the generated asset can also be seamlessly integrated into downstream manipulation pipelines, e.g. animation, composition, and simulation etc., greatly promoting its potential in wide applications. |
 |
Shuai Zhang, Huangxuan Zhao, Zhenghong Zhou, Guanjun Wu, Chuansheng Zheng, Wenyu Liu, Xinggang Wang, IEEE JBHI [ Paper] [ Code] We propose TOGS, a Gaussian splatting method with opacity offset over time, which can effectively improve the rendering quality and speed of 4D DSA. We introduce an opacity offset table for each Gaussian to model the temporal variations in the radiance of the contrast agent. Additionally, we introduced a Smooth loss term in the loss function to mitigate overfitting issues that may arise in the model when dealing with sparse view scenarios. This model achieves stateof-the-art reconstruction quality under the same number of training views. Additionally, it enables real-time rendering while maintaining low storage overhead. |
 |
Taoran Yi, Jiemin Fang, Junjie Wang, Guanjun Wu, Lingxi Xie, Xiaopeng Zhang, Wenyu Liu, Qi Tian, Xinggang Wang CVPR, 2024 [ Paper] [ Page] [ Code] A fast 3D generation framework, named as GaussianDreamer, is proposed, where the 3D diffusion model provides point cloud priors for initialization and the 2D diffusion model enriches the geometry and appearance. Our GaussianDreamer can generate a high-quality 3D instance within 25 minutes on one GPU, much faster than previous methods, while the generated instances can be directly rendered in real time. |
|
|
|
I really like playing games, including but not limited to:
|
|
Template is adapted from
Here
|