Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Unsupervised Learning for Cuboid Shape Abstraction via Joint Segmentation from Point Clouds

About

Representing complex 3D objects as simple geometric primitives, known as shape abstraction, is important for geometric modeling, structural analysis, and shape synthesis. In this paper, we propose an unsupervised shape abstraction method to map a point cloud into a compact cuboid representation. We jointly predict cuboid allocation as part segmentation and cuboid shapes and enforce the consistency between the segmentation and shape abstraction for self-learning. For the cuboid abstraction task, we transform the input point cloud into a set of parametric cuboids using a variational auto-encoder network. The segmentation network allocates each point into a cuboid considering the point-cuboid affinity. Without manual annotations of parts in point clouds, we design four novel losses to jointly supervise the two branches in terms of geometric similarity and cuboid compactness. We evaluate our method on multiple shape collections and demonstrate its superiority over existing shape abstraction methods. Moreover, based on our network architecture and learned representations, our approach supports various applications including structured shape generation, shape interpolation, and structural shape clustering.

Kaizhi Yang, Xuejin Chen• 2021

Related benchmarks

TaskDatasetResultRank
3D Building AbstractionSparse Samping Point Cloud
Vertex Count56
6
3D Building AbstractionSfM Point Cloud
#V (Vertices)60
6
3D Building AbstractionUser Study
User Study Score0.1
6
3D Object ReconstructionScanNet++ out-of-category 58
L1 Chamfer Distance2.91
4
3D Object ReconstructionReplica 45 (out-of-category)
L1 Chamfer Distance3.68
4
Object DecompositionShapeNet In-category (test)
L1 Error5.157
4
Object DecompositionShapeNet Out-of-category (test)
L1 Error4.897
4
Showing 7 of 7 rows

Other info

Follow for update