The scarcity of annotated data in outdoor point cloud segmentation poses a
significant obstacle in harnessing the modeling capabilities of advanced
networks like transformers. Consequently, scholars have been actively
investigating efficacious self-supervised pre-training strategies, e.g.
contrasting learning and reconstruction-based pretext tasks. Nevertheless,
temporal information, which is inherent in the LiDAR point cloud sequence, is
consistently disregarded. To better utilize this property, we propose an
effective pre-training strategy, namely Temporal Masked AutoEncoders (T-MAE),
which takes as input temporally adjacent frames and learns temporal dependency.
A SiamWCA backbone, containing a Siamese encoder and a window-based
cross-attention (WCA) module, is established for the two-frame input. Taking
into account that the motion of an ego-vehicle alters the illumination angles
of the same instance, temporal modeling also serves as a robust and natural
data augmentation, enhancing the comprehension of target objects. Moreover,
instead of utilizing consecutive frames, it is more cost-effective and powerful
by using distant historical frames. SiamWCA is a powerful architecture but
heavily relies on annotated data. With our T-MAE pre-training strategy, we
achieve the best performance on the Waymo dataset among self-supervised
learning methods. Comprehensive experiments are conducted to validate all
components of our proposal. Upon acceptance, the source code will be made
accessible.

在户外点云分割中，基于转换器等先进网络的建模能力受制于标注数据的稀缺性。为了更好地利用点云序列中蕴含的时间信息，本文提出了一种有效的预训练策略，即时序遮蔽自动编码器 (Temporal Masked AutoEncoders, T-MAE)，通过输入与时间相关的相邻帧并学习时间依赖关系。T-MAE 预训练策略结合了 SiamWCA 骨干网络，其中包含了孪生编码器和基于窗口的交叉注意力模块 (window-based cross-attention, WCA)，用于接收两帧输入。此外，为了提高目标对象的理解能力，时间建模还作为一种稳健且天然的数据增强方法，在同一对象的不同帧之间考虑自车运动改变的照明角度。通过使用遥远的历史帧而非连续帧，可以更加成本有效且强大。在自监督学习方法中，利用 T-MAE 预训练策略在 Waymo 数据集上实现了最佳性能。通过全面实验验证了本提案的所有组成部分。在接受后，源代码将会公开。