比较了多种3D旋转表示法对深度学习性能的影响。
On Representation of 3D Rotation in the Context of Deep Learning
- 采用连续的5D和6D表示法提升旋转估计精度
- 在合成与真实数据上均验证了连续表示更优
- 适合从事3D视觉与神经网络研究者参考
本文研究了多种3D旋转表示方法对深度神经网络学习过程的影响。我们在合成与真实数据上评估了ResNet18网络在不同旋转表示和损失函数下的3D旋转估计性能。真实数据包含工业料箱的3D扫描，合成数据则为简单非对称物体在不同旋转视角下的渲染图像。在合成数据上，还分析了训练与测试集中的旋转分布差异及物体纹理的影响。结果表明，与不连续表示相比，使用连续的5D和6D表示法的网络表现更优，与先前研究一致。
原文摘要 · Abstract (English)
This paper investigates various methods of representing 3D rotations and their impact on the learning process of deep neural networks. We evaluated the performance of ResNet18 networks for 3D rotation estimation using several rotation representations and loss functions on both synthetic and real data. The real datasets contained 3D scans of industrial bins, while the synthetic datasets included views of a simple asymmetric object rendered under different rotations. On synthetic data, we also assessed the effects of different rotation distributions within the training and test sets, as well as the impact of the object's texture. In line with previous research, we found that networks using the continuous 5D and 6D representations performed better than the discontinuous ones.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。