In this paper, we propose SelfSphNet, that is, a self-supervised learning network to estimate the motion of an arbitrarily moving spherical camera without the need for any labeled training data. Recently, numerous learning-based methods for camera motion estimation have been proposed. However, most of these methods require an enormous amount of labeled training data, which is difficult to acquire experimentally. To solve this problem, our SelfSphNet employs two loss functions to estimate the frame-to-frame camera motion, thus giving two supervision signals to the network with the usage of unlabeled training data. First, a 5 DoF epipolar angular loss, which is composed of a dense optical flow of spherical images, estimates the 5 DoF motion between two image frames. This loss function utilizes a unique property of the spherical optical flow, which allows the rotational and translational components to be decoupled by using a derotation operation. This operation is derived from the fact that spherical images can be rotated to any orientation without any loss of information, hence making it possible to ''decouple'' the dense optical flow between pairs of spherical images to a pure translational state. Next, a photometric reprojection loss estimates the full 6 DoF motion using a depth map generated from the decoupled optical flow. This minimization strategy enables our network to be optimized without using any labeled training data. To confirm the effectiveness of our proposed approach (SelfSphNet), several experiments to estimate the camera trajectory, as well as the camera motion, were conducted in comparison to a previous self-supervised learning approach, SfMLearner, and a fully supervised learning approach whose baseline network is the same as SelfSphNet. Moreover, transfer learning in a new scene was also conducted to verify that our proposed method can optimize the network with newly collected unlabeled data. INDEX TERMS Motion estimation, computer vision, image processing, deep learning, convolutional neural networks.
In this paper, we present a novel method for caging-based grasping of deformable objects. This method enables manipulators to grasp objects simply with geometric constraints by using position control of robotic hands, and not through force controls or mechanical analysis. Therefore, this method has cost benefits and algorithmic simplicity. In our previous studies, we mainly focused on caging-based grasping of rigid objects such as 2D/3D primitive-shaped objects. However, considering realistic objects, manipulation of deformable objects is also required frequently. Hence, this study is motivated to manipulate deformable objects, adopting a caging-based grasping approach. We formulate caging-based grasping of deformable objects, and target three types of deformable objects: a rigid object covered with a soft part, a closed-loop structure, and two rigid bodies connected with a string, which can be regarded as primitive shapes. We then derive concrete conditions for grasp synthesis and conduct experimental verification of our proposed method with an industrial manipulator.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.