跳到正文
返回论文列表
cs.GR提交于 已译

SemEllipsoid:语义感知的椭球形状抽象

SemEllipsoid: Semantic-aware Shape Abstraction with Ellipsoid

Kuankuan Cheng · Haocheng Yuan · Changjian Li

中文摘要

用一组基元表示三维形状(形状抽象)是计算机图形学中活跃的研究问题,因基元表示紧凑且计算高效,在形状理解、分析与仿真中应用广泛。已有工作主要关注几何感知的形状抽象,目标是用最少基元逼近输入形状并保持几何保真度,但常忽略语义一致性,导致基元跨越多个语义部件,在下游的部件级形状编辑或操作中产生不良伪影。本文提出新任务——语义感知的形状抽象,目标是生成既几何保真又语义一致的抽象,确保每个基元严格属于单一语义部件、不跨越语义边界。为此提出 SemEllipsoid,使用椭球基元解决该问题。核心思想是利用大规模基础模型提供的语义线索,将语义约束注入基元优化过程。设计全可微的投影-混合-评估方案,以输入语义掩码为参照度量语义偏差,并构建交替优化策略,迭代求解最优语义分配与基元配置。该方法非学习式,无需数据收集或网络训练,可泛化到任意类别形状,并支持按用户需求调整抽象层级。在多个三维形状集合上的实验与评估验证了有效性。

关键要点

  1. 01已有几何感知形状抽象忽略语义一致性,基元常跨越多个语义部件,导致下游编辑出现伪影。
  2. 02提出语义感知形状抽象新任务,要求基元几何保真且严格不跨越语义边界。
  3. 03SemEllipsoid 利用基础模型语义线索,以全可微投影-混合-评估度量语义偏差并交替优化。
  4. 04方法非学习式,无需训练数据,可泛化到任意类别并支持多层级抽象。
  5. 05在多个三维形状集合上实验验证了方法的有效性。

解读

尚无解读。

原始英文摘要

arXiv:2610.42145v1 Announce Type: cross Abstract: Representing 3D shapes using a set of primitives - known as shape abstraction - is an actively studied problem in computer graphics with wide applications in shape understanding, analysis, and simulation due to the compactness and computational efficiency of the primitive-based representation. Prior work has primarily focused on geometry-aware shape abstraction, where the objective is to approximate the input shape with a minimal set of primitives while preserving geometric fidelity. However, these methods often overlook semantic consistency, inevitably leading to primitives that span across multiple semantic parts. Such inconsistency can cause undesirable artifacts in downstream tasks such as part-level shape editing or manipulation. In this paper, we introduce a new task - semantic-aware shape abstraction, where the goal is to produce abstractions that are both geometrically faithful and semantically consistent, ensuring that each primitive adheres strictly to a single semantic part without crossing semantic boundaries. We propose SemEllipsoid, a novel method to address this problem using ellipsoidal primitives. The core idea is to inject semantic constraints into the primitive optimization process leveraging the availability of semantic cues from large-scale foundation models. We design a fully differentiable projection-blending-evaluation scheme to measure semantic divergence against input semantic masks, and formulate an alternating optimization strategy to iteratively solve for optimal semantic assignment and primitive configurations. Our method is non-learning-based, requiring no data collection or network training, and generalizes well to shapes from arbitrary categories. It also supports different levels of abstraction according to user demand. Extensive experiments and evaluations validate the effectiveness of our approach across several 3D shape collections.

同方向论文 · cs.GR

查看全部 →