SweepNet: Unsupervised Learning Shape Abstraction via Neural Sweepers

ECCV 2024

Mingrui Zhao¹, Yizhi Wang¹, Fenggen Yu¹, Changqing Zou², Ali Mahdavi-Amiri¹,
¹Simon Fraser University ²Zhejiang University

Paper

Code

Dataset (huggingface)

Dataset (Onedrive)

Abstract

Hello! This is SweepNet. Each letter is represented with parametrisable sweep surfaces.

Shape abstraction is an important task for simplifying complex geometric structures while retaining essential features. Sweep surfaces, commonly found in human-made objects, aid in this process by effectively capturing and representing object geometry, thereby facilitating abstraction. In this paper, we introduce \papername, a novel approach to shape abstraction through sweep surfaces. We propose an effective parameterization for sweep surfaces, utilizing superellipses for profile representation and B-spline curves for the axis. This compact representation, requiring as few as 14 float numbers, facilitates intuitive and interactive editing while preserving shape details effectively. Additionally, by introducing a differentiable neural sweeper and an encoder-decoder architecture, we demonstrate the ability to predict sweep surface representations without supervision. We show the superiority of our model through several quantitative and qualitative experiments throughout the paper.

Method

Pipeline overview. The model processes voxel input to extract a skeletal prior and encodes the data with a voxel encoder. The sweep surface head predicts sweep surface parameters: 2D profiles, 3D sweeping axes, and profile scale coefficients, conditioned on the skeletal prior. Training involves generating key point clouds for each sweep surface through a differentiable sampler, which the neural sweeper uses to estimate sweep surface occupancy. This data is then assembled to reconstruct the input shape to quantify loss. At inference time, the sweep surface parameters are directly processed by a non-differentiable, imperative sweeper to produce the resembled shape.

Primitive Parametrisation

Linear sweep surface

+Scaling function

Curvy sweep surface

+Scaling function

Sweep surface primitives are created by sweeping a 2D profile along a 3D axis, with a scaling function controls profile scale during the sweeping process.