Underwater light field retention: Neural rendering for underwater imaging

T Ye, S Chen, Y Liu, Y Ye… - Proceedings of the …, 2022 - openaccess.thecvf.com
Abstract Underwater Image Rendering aims to generate a true-to-life underwater image from
a given clean one, which could be applied to various practical applications such as …

Lart: Neural correspondence learning with latent regularization transformer for 3d motion transfer

H Chen, H Tang, R Timofte… - Advances in Neural …, 2024 - proceedings.neurips.cc
Abstract 3D motion transfer aims at transferring the motion from a dynamic input sequence to
a static 3D object and outputs an identical motion of the target with high-fidelity and realistic …

Unsupervised 3d pose transfer with cross consistency and dual reconstruction

C Song, J Wei, R Li, F Liu, G Lin - IEEE Transactions on Pattern …, 2023 - ieeexplore.ieee.org
The goal of 3D pose transfer is to transfer the pose from the source mesh to the target mesh
while preserving the identity information (eg, face, body shape) of the target mesh. Deep …

Weakly-supervised 3d pose transfer with keypoints

J Chen, C Li, GH Lee - Proceedings of the IEEE/CVF …, 2023 - openaccess.thecvf.com
The main challenges of 3D pose transfer are: 1) Lack of paired training data with different
characters performing the same pose; 2) Disentangling pose and shape information from the …

Geometry-contrastive transformer for generalized 3d pose transfer

H Chen, H Tang, Z Yu, N Sebe, G Zhao - Proceedings of the AAAI …, 2022 - ojs.aaai.org
We present a customized 3D mesh Transformer model for the pose transfer task. As the 3D
pose transfer essentially is a deformation procedure dependent on the given meshes, the …

Zero-shot pose transfer for unrigged stylized 3d characters

J Wang, X Li, S Liu, S De Mello… - Proceedings of the …, 2023 - openaccess.thecvf.com
Transferring the pose of a reference avatar to stylized 3D characters of various shapes is a
fundamental task in computer graphics. Existing methods either require the stylized …

Shapefusion: A 3d diffusion model for localized shape editing

RA Potamias, M Tarasiou, S Ploumpis… - European Conference on …, 2025 - Springer
In the realm of 3D computer vision, parametric models have emerged as a ground-breaking
methodology for the creation of realistic and expressive 3D avatars. Traditionally, they rely …

Towards Robust 3D Pose Transfer with Adversarial Learning

H Chen, H Tang, E Adeli… - Proceedings of the IEEE …, 2024 - openaccess.thecvf.com
Abstract 3D pose transfer that aims to transfer the desired pose to a target mesh is one of the
most challenging 3D generation tasks. Previous attempts rely on well-defined parametric …

Aniformer: Data-driven 3d animation with transformer

H Chen, H Tang, N Sebe, G Zhao - arXiv preprint arXiv:2110.10533, 2021 - arxiv.org
We present a novel task, ie, animating a target 3D object through the motion of a raw driving
sequence. In previous works, extra auxiliary correlations between source and target meshes …

3d generative model latent disentanglement via local eigenprojection

S Foti, B Koo, D Stoyanov… - Computer Graphics …, 2023 - Wiley Online Library
Designing realistic digital humans is extremely complex. Most data‐driven generative
models used to simplify the creation of their underlying geometric shape do not offer control …