Bevcontrast: Self-supervision in bev space for automotive lidar point clouds

C Sautier, G Puy, A Boulch, R Marlet… - … Conference on 3D …, 2024 - ieeexplore.ieee.org
We present a surprisingly simple and efficient method for self-supervision of 3D backbone
on automotive Lidar point clouds. We design a contrastive loss between features of Lidar …

Parameter Efficient Point Cloud Prompt Tuning for Unified Point Cloud Understanding

B Fei, L Liu, W Yang, Z Li, WM Chen… - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
The prevailing paradigm in 3D vision involves fully fine-tuning all the backbone parameters
of pre-trained models. However, this approach poses challenges due to the large number of …

A self-supervised learning network for student engagement recognition from facial expressions

WL Zhang, RS Jia, H Wang, CY Che… - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
Student engagement in online learning is an important indicator for measuring learning
effectiveness. Due to the fact that facial video data of students during online learning …

Exploring Self-Supervised Learning for 3D Point Cloud Registration

M Yuan, Q Huang, A Shen, X Huang… - IEEE Robotics and …, 2024 - ieeexplore.ieee.org
Self-supervised learning has achieved significant success in various fields such as point
cloud detection and segmentation. However, self-supervised learning for point cloud …

Complementary multi-modality molecular self-supervised learning via non-overlapping masking for property prediction

A Shen, M Yuan, Y Ma, J Du… - Briefings in …, 2024 - academic.oup.com
Self-supervised learning plays an important role in molecular representation learning
because labeled molecular data are usually limited in many tasks, such as chemical …

PatchMixing Masked Autoencoders for 3D Point Cloud Self-Supervised Learning

C Lin, W Xu, J Zhu, Y Nie, R Cai… - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
Recently, Point-MAE has extended Masked Autoencoders (MAE) to point clouds for 3D self-
supervised learning, which however faces two problems:(1) the shape similarity between the …

Cross-BERT for Point Cloud Pretraining

X Li, P Li, Z Wei, Z Zhu, M Wei, J Hou, L Nan… - arXiv preprint arXiv …, 2023 - arxiv.org
Introducing BERT into cross-modal settings raises difficulties in its optimization for handling
multiple modalities. Both the BERT architecture and training objective need to be adapted to …