UprightRL: Upright Orientation Estimation of 3D Shapes via Reinforcement Learning

Simple item page
dc.contributor.authorChen, Luanminen_US
dc.contributor.authorXu, Juzhanen_US
dc.contributor.authorWang, Chuanen_US
dc.contributor.authorHuang, Haibinen_US
dc.contributor.authorHuang, Huien_US
dc.contributor.authorHu, Ruizhenen_US
dc.contributor.editorZhang, Fang-Lue and Eisemann, Elmar and Singh, Karanen_US
dc.date.accessioned2021-10-14T11:12:24Z
dc.date.available2021-10-14T11:12:24Z
dc.date.issued2021
dc.description.abstractIn this paper, we study the problem of 3D shape upright orientation estimation from the perspective of reinforcement learning, i.e. we teach a machine (agent) to orientate 3D shapes step by step to upright given its current observation. Unlike previous methods, we take this problem as a sequential decision-making process instead of a strong supervised learning problem. To achieve this, we propose UprightRL, a deep network architecture designed for upright orientation estimation. UprightRL mainly consists of two submodules: an Actor module and a Critic module which can be learned with a reinforcement learning manner. Specifically, the Actor module selects an action from the action space to perform a point cloud transformation and obtain the new point cloud for the next environment state, while the Critic module evaluates the strategy and guides the Actor to choose the next stage action. Moreover, we design a reward function that encourages the agent to select action which is conducive to orient model towards upright orientation with a positive reward and negative otherwise. We conducted extensive experiments to demonstrate the effectiveness of the proposed model, and experimental results show that our network outperforms the stateof- the-art. We also apply our method to the robot grasping-and-placing experiment, to reveal the practicability of our method.en_US
dc.description.number7
dc.description.sectionheadersModelling
dc.description.seriesinformationComputer Graphics Forum
dc.description.volume40
dc.identifier.doi10.1111/cgf.14419
dc.identifier.issn1467-8659
dc.identifier.pages265-275
dc.identifier.urihttps://doi.org/10.1111/cgf.14419
dc.identifier.urihttps://diglib.eg.org:443/handle/10.1111/cgf14419
dc.publisherThe Eurographics Association and John Wiley & Sons Ltd.en_US
dc.subjectComputing methodologies
dc.subjectReinforcement learning
dc.subjectSequential decision
dc.subjectmaking process
dc.subjectUpright orientation
dc.titleUprightRL: Upright Orientation Estimation of 3D Shapes via Reinforcement Learningen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
v40i7pp265-275.pdf
Size:
11.74 MB
Format:
Adobe Portable Document Format
Download
Collections
40-Issue 7