ASM-Net: Category-level Pose and Shape Estimation Using Parametric Deformation
ASM-Net: Category-level Pose and Shape Estimation Using Parametric Deformation
复制标题
DOI：
--
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Shuichi Akizuki;M. Hashimoto
中科院分区：
文献类型：
--
作者：
Shuichi Akizuki;M. Hashimoto
我们提出了一种新的深度神经网络，它可以从点云数据中估计六个自由度的姿态和看不见的物体的完整形状。我们的概念是只使用目标类别的3D模型来训练网络，该网络可以在消费者RGBD相机捕获的真实的图像上表现良好。为此，我们采用了两种想法。第一种是使用活动形状模型对类别内形状变化进行建模，活动形状模型可以用几个维度参数使形状变形。第二个是对训练数据应用有效的过滤过程，将3D对象模型转换为模拟传感器测量的点云。我们在NOCS REAL275上评估了我们的方法，NOCS REAL275是一个广泛使用的用于类别级姿态估计的基准数据集，并证实了其在形状恢复和姿态估计方面优于传统方法。我们的代码可在https://github.com/sakizuki/asm-net上获得。
We propose a novel deep neural network that estimates the six degrees of freedom pose and complete shape of unseen objects from point cloud data. Our concept is to train the network that can perform well on real images captured by a consumer RGBD camera using only 3D models of the target category. To do so, we have employed two ideas. The first is modeling intra-category shape variations with active shape models that can deform the shape with a few dimensional parameters. The second is applying effective filtering processes to the training data to convert the 3D object model into a point cloud that simulates the sensor measurements. We evaluated our method on NOCS REAL275, a widely used benchmark dataset for category-level pose estimation, and confirmed its superiority over conventional methods in terms of both shape recovery and pose estimation. Our code is available at https://github.com/sakizuki/asm-net .