Bevcontrast: Self-supervision in bev space for automotive lidar point clouds
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 …
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
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 …
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 …
effectiveness. Due to the fact that facial video data of students during online learning …
Exploring Self-Supervised Learning for 3D Point Cloud Registration
Self-supervised learning has achieved significant success in various fields such as point
cloud detection and segmentation. However, self-supervised learning for point cloud …
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
Self-supervised learning plays an important role in molecular representation learning
because labeled molecular data are usually limited in many tasks, such as chemical …
because labeled molecular data are usually limited in many tasks, such as chemical …
PatchMixing Masked Autoencoders for 3D Point Cloud Self-Supervised Learning
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 …
supervised learning, which however faces two problems:(1) the shape similarity between the …
Cross-BERT for Point Cloud Pretraining
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 …
multiple modalities. Both the BERT architecture and training objective need to be adapted to …