CFPNet：通过跨区域特征传播改进轻量化飞行时间深度补全
CFPNet: Improving Lightweight ToF Depth Completion via Cross-zone Feature Propagation
AI总结：
针对轻量化 ToF 传感器视场有限导致区域外深度补全性能下降的问题，本文提出 CFPNet，利用 DAPM 与 LKPM 实现跨区域特征传播，并在 ZJU-L5 上达到 SOTA。
摘要：
使用轻量化飞行时间（ToF）深度传感器进行深度补全因其低成本而具有吸引力。然而，与相机相比，轻量化 ToF 传感器通常具有有限的视场（FOV）。因此，只有图像区域内的像素能够与深度信号相关联。以往方法未能有效地将深度特征从区域内传播到区域外，因此在区域外的深度补全性能会下降。为此，本文提出 CFPNet，通过两个新颖模块实现从区域内到区域外的跨区域特征传播。第一个是基于直接注意力的传播模块（DAPM），它强制实现直接的跨区域特征获取。第二个是基于大核的传播模块（LKPM），它通过利用核大小最大为 31 的卷积层实现跨区域特征传播。通过恰当地结合这两个模块，CFPNet 取得了最先进（SOTA）的深度补全性能，这一点已由在 ZJU-L5 数据集上的大量实验结果验证。代码可在 https://github.com/denyingmxd/CFPNet 获取。
英文摘要：
Depth completion using lightweight time-of-flight (ToF) depth sensors is attractive due to their low cost. However, lightweight ToF sensors usually have a limited field of view (FOV) compared with cameras. Thus, only pixels in the zone area of the image can be associated with depth signals. Previous methods fail to propagate depth features from the zone area to the outside-zone area effectively, thus suffering from degraded depth completion performance outside the zone. To this end, this paper proposes the CFPNet to achieve cross-zone feature propagation from the zone area to the outside-zone area with two novel modules. The first is a direct-attention-based propagation module (DAPM), which enforces direct cross-zone feature acquisition. The second is a large-kernel-based propagation module (LKPM), which realizes cross-zone feature propagation by utilizing convolution layers with kernel sizes up to 31. CFPNet achieves state-of-the-art (SOTA) depth completion performance by combining these two modules properly, as verified by extensive experimental results on the ZJU-L5 dataset. The code is available at https://github.com/denyingmxd/CFPNet.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
