arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2209.09231cs.CV
收藏

3D- PL:具有3D感知伪标记的域自适应深度估计

3D-PL: Domain Adaptive Depth Estimation with 3D-aware Pseudo-Labeling

National Chiao Tung University(国立交通大学);Phiar Technologies(Phiar科技公司)

Yu-Ting Yen, Chia-Ni Lu, Wei-Chen Chiu, Yi-Hsuan Tsai

摘要:

对于单目深度估计,获取真实数据的地面实况并不容易,因此通常采用使用监督合成数据的领域自适应方法。然而,由于缺乏来自真实数据的监督,这仍然可能导致很大的域缺口。在本文中,我们开发了一个领域自适应框架,通过生成可靠的伪地面实况的深度从实际数据提供直接的监督。具体来说,我们提出了两种伪标签机制:1)基于2D的伪标签,通过测量图像内容相同但风格不同时深度预测的一致性; 2)3D感知的伪标签,通过点云完成网络学习在3D空间中完成深度值,从而在场景中提供更多的结构信息,以改进和生成更可靠的伪标签。在实验中,我们证明了我们的伪标记方法在各种设置中改善了深度估计,包括在训练期间使用立体声对。此外,所提出的方法在现实世界的数据集上表现良好,对几个国家的最先进的无监督领域自适应方法。

英文摘要:

For monocular depth estimation, acquiring ground truths for real data is not easy, and thus domain adaptation methods are commonly adopted using the supervised synthetic data. However, this may still incur a large domain gap due to the lack of supervision from the real data. In this paper, we develop a domain adaptation framework via generating reliable pseudo ground truths of depth from real data to provide direct supervisions. Specifically, we propose two mechanisms for pseudo-labeling: 1) 2D-based pseudo-labels via measuring the consistency of depth predictions when images are with the same content but different styles; 2) 3D-aware pseudo-labels via a point cloud completion network that learns to complete the depth values in the 3D space, thus providing more structural information in a scene to refine and generate more reliable pseudo-labels. In experiments, we show that our pseudo-labeling methods improve depth estimation in various settings, including the usage of stereo pairs during training. Furthermore, the proposed method performs favorably against several state-of-the-art unsupervised domain adaptation approaches in real-world datasets.

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。

↑