Puzzleavatar: Assembling 3d avatars from personal albums

Y Xiu, Y Ye, Z Liu, D Tzionas, MJ Black - ACM Transactions on Graphics …, 2024 - dl.acm.org
Generating personalized 3D avatars is crucial for AR/VR. However, recent text-to-3D
methods that generate avatars for celebrities or fictional characters, struggle with everyday …

Training-free model merging for multi-target domain adaptation

W Li, H Gao, M Gao, B Tian, R Zhi, H Zhao - European Conference on …, 2025 - Springer
In this paper, we study multi-target domain adaptation of scene understanding models.
While previous methods achieved commendable results through inter-domain consistency …

Innovative AI techniques for photorealistic 3D clothed human reconstruction from monocular images or videos: a survey

S Yang, X Gu, Z Kuang, F Qin, Z Wu - The Visual Computer, 2024 - Springer
The reconstruction of high-quality 3D clothed humans from monocular images or videos has
gained popularity in recent years due to its significant practical applications. While several …

SCP-Diff: Spatial-Categorical Joint Prior for Diffusion Based Semantic Image Synthesis

H Gao, M Gao, J Li, W Li, R Zhi, H Tang… - European Conference on …, 2025 - Springer
Semantic image synthesis (SIS) shows good promises for sensor simulation. However,
current best practices in this field, based on GANs, have not yet reached the desired level of …

DRiVE: Diffusion-based Rigging Empowers Generation of Versatile and Expressive Characters

M Sun, J Chen, J Dong, Y Chen, X Jiang, S Mao… - arXiv preprint arXiv …, 2024 - arxiv.org
Recent advances in generative models have enabled high-quality 3D character
reconstruction from multi-modal. However, animating these generated characters remains a …

AniGS: Animatable Gaussian Avatar from a Single Image with Inconsistent Gaussian Reconstruction

L Qiu, S Zhu, Q Zuo, X Gu, Y Dong, J Zhang… - arXiv preprint arXiv …, 2024 - arxiv.org
Generating animatable human avatars from a single image is essential for various digital
human modeling applications. Existing 3D reconstruction methods often struggle to capture …

[HTML][HTML] Enhanced Multi-Scale Attention-Driven 3D Human Reconstruction from Single Image

Y Ren, M Zhou, P Zhou, S Wang, Y Liu, G Geng, K Li… - Electronics, 2024 - mdpi.com
Due to the inherent limitations of a single viewpoint, reconstructing 3D human meshes from
a single image has long been a challenging task. While deep learning networks enable us …

Diffusion-based Visual Anagram as Multi-task Learning

Z Xu, Y Chen, H Gao, W Zhao, G Zhang… - arXiv preprint arXiv …, 2024 - arxiv.org
Visual anagrams are images that change appearance upon transformation, like flipping or
rotation. With the advent of diffusion models, generating such optical illusions can be …

Generating View-Changing Human Action Videos Using a Single Reference Image and Textual Prompting

JSK Akula - 2024 - search.proquest.com
Human motion synthesis involves generating videos of a human performing specified
actions from a reference image through pose inputs. Applications such as fitness and dance …