Relationship-Based Point Cloud Completion
- 3 September 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Visualization and Computer Graphics
- Vol. 28 (12) , 4940-4950
Abstract
We propose a partial point cloud completion approach for scenes that are composed of multiple objects. We focus on pairwise scenes where two objects are in close proximity and are contextually related to each other, such as a chair tucked in a desk, a fruit in a basket, a hat on a hook and a flower in a vase. Different from existing point cloud completion methods, which mainly focus on single objects, we design a network that encodes not only the geometry of the individual shapes, but also the spatial relations between different objects. More specifically, we complete missing parts of the objects in a conditional manner, where the partial or completed point cloud of the other object is used as an additional input to help predict missing parts. Based on the idea of conditional completion, we further propose a two-path network, which is guided by a consistency loss between different sequences of completion. Our method can handle difficult cases where the objects heavily occlude each other. Also, it only requires a small set of training data to reconstruct the interaction area compared to existing completion approaches. We evaluate our method qualitatively and quantitatively via ablation studies and in comparison to the state-of-the-art point cloud completion methods.Keywords
Funding Information
- National Natural Science Foundation of China (62072366, 61872250)
- China Postdoctoral Science Foundation (2020M673407)
- Fundamental Research Funds for the Central Universities (xzy012019048)
- Guangdong Natural Science Foundation (2021B1515020085)
- University of Hong Kong (182DRTAKU, 187FRTAKU, 230DRTAKU)
This publication has 30 references indexed in Scilit:
- PointNet: A 3D Convolutional Neural Network for real-time object class recognition2016
- Deep Sliding Shapes for Amodal 3D Object Detection in RGB-D Images2016
- Consistent Partial Matching of Shape Collections via Sparse ModelingComputer Graphics Forum, 2016
- Interaction context (ICON)ACM Transactions on Graphics, 2015
- Functional map networks for analyzing and exploring large shape collectionsACM Transactions on Graphics, 2014
- Indexing 3D Scenes Using the Interaction Bisector SurfaceACM Transactions on Graphics, 2014
- Sliding Shapes for 3D Object Detection in Depth Images2014
- Consistent Shape Maps via Semidefinite ProgrammingComputer Graphics Forum, 2013
- A Correlated Parts Model for Object Detection in Large 3D ScansComputer Graphics Forum, 2013
- Example-based synthesis of 3D object arrangementsACM Transactions on Graphics, 2012