A Survey on Self-supervised Learning: Algorithms, Applications, and Future Trends

J Gui, T Chen, J Zhang, Q Cao, Z Sun… - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
Deep supervised learning algorithms typically require a large volume of labeled data to
achieve satisfactory performance. However, the process of collecting and labeling such data …

Masked modeling for self-supervised representation learning on vision and beyond

S Li, L Zhang, Z Wang, D Wu, L Wu, Z Liu, J **a… - arxiv preprint arxiv …, 2023 - arxiv.org
As the deep learning revolution marches on, self-supervised learning has garnered
increasing attention in recent years thanks to its remarkable representation learning ability …

Nerf-mae: Masked autoencoders for self-supervised 3d representation learning for neural radiance fields

MZ Irshad, S Zakharov, V Guizilini, A Gaidon… - … on Computer Vision, 2024 - Springer
Neural fields excel in computer vision and robotics due to their ability to understand the 3D
visual world such as inferring semantics, geometry, and dynamics. Given the capabilities of …

Fac: 3d representation learning via foreground aware feature contrast

K Liu, A **ao, X Zhang, S Lu… - Proceedings of the IEEE …, 2023 - openaccess.thecvf.com
Contrastive learning has recently demonstrated great potential for unsupervised pre-training
in 3D scene understanding tasks. However, most existing work randomly selects point …

MeshCLIP: Efficient cross-modal information processing for 3D mesh data in zero/few-shot learning

Y Song, N Liang, Q Guo, J Dai, J Bai, F He - Information Processing & …, 2023 - Elsevier
Abstract Text, 2D, and 3D information are crucial information representations in modern
science and management disciplines. However, complex and irregular 3D data produce …

A survey on self-supervised learning: Algorithms, applications, and future trends

J Gui, T Chen, J Zhang, Q Cao, Z Sun, H Luo… - arxiv preprint arxiv …, 2023 - arxiv.org
Deep supervised learning algorithms typically require a large volume of labeled data to
achieve satisfactory performance. However, the process of collecting and labeling such data …

Unsupervised 3D skeleton-based action recognition using cross-attention with conditioned generation capabilities

DJ Lerch, Z Zhong, M Martin, M Voit… - Proceedings of the …, 2024 - openaccess.thecvf.com
Human action recognition plays a pivotal role in various real-world applications, including
surveillance systems, robotics, and occupant monitoring in the car interior. With such a …

Mesh neural networks based on dual graph pyramids

XL Li, ZN Liu, T Chen, TJ Mu… - IEEE Transactions on …, 2023 - ieeexplore.ieee.org
Deep neural networks (DNNs) have been widely used for mesh processing in recent years.
However, current DNNs can not process arbitrary meshes efficiently. On the one hand, most …

Self-Supervised Learning with Masked Autoencoders for Teeth Segmentation from Intra-oral 3D Scans

A Almalki, LJ Latecki - Proceedings of the IEEE/CVF Winter …, 2024 - openaccess.thecvf.com
In modern dentistry, teeth localization, segmentation, and labeling from intra-oral 3D scans
are crucial for improving dental diagnostics, treatment planning, and population-based …

Shapesplat: A large-scale dataset of gaussian splats and their self-supervised pretraining

Q Ma, Y Li, B Ren, N Sebe, E Konukoglu… - arxiv preprint arxiv …, 2024 - arxiv.org
3D Gaussian Splatting (3DGS) has become the de facto method of 3D representation in
many vision tasks. This calls for the 3D understanding directly in this representation space …