给3D分割模型加个轻量几何先验层，提升精度不增负担
GIBLy: Improving 3D Semantic Segmentation through an Architecture-Agnostic Lightweight Geometric Inductive Bias Layer
- 插入可学习的几何形状先验，适配各类3D网络架构
- 在TS40K上提升11.5% mIoU，仅增加5.8万参数
- 适合追求高精度且受限于算力的3D场景理解任务
在3D场景理解中，深度学习模型依赖大模型和大量训练来捕捉3D数据中的基本几何结构。然而，现有方法缺乏显式机制融入几何信息（如可学习的原始形状），常需大模型和更多训练数据，导致成本上升并限制泛化能力。本文提出GIBLy，一种架构无关的轻量级几何归纳偏置层，将可学习的几何先验融入3D分割流程。GIBLy通过提供与简单几何形状对齐的特征（具有人类可解释性），在不增加计算开销的前提下，增强现有架构（无论基于MLP、卷积或Transformer）的分割性能。我们在多个3D语义分割基准上验证该方法，表现持续提升，包括在TS40K上使用PTV3时实现+11.5% mIoU，且仅引入58,000额外参数。结果表明，显式编码几何结构有助于实现准确高效的3D场景理解。
原文摘要 · Abstract (English)
In 3D scene understanding, deep learning models rely on large models and extensive training to capture basic geometric structures that are present in the 3D data. However, existing methods lack explicit mechanisms to incorporate geometric information, such as learnable primitive shapes, often necessitating large models and more training data which in turn increases cost and can limit generalization. We introduce GIBLy, a lightweight geometric inductive bias layer that integrates learnable geometric priors into 3D segmentation pipelines. GIBLy enhances existing architectures -- whether MLP-based, convolution-based, or transformer-based -- by providing features aligned with simple geometric shapes (and thus human-interpretable) that improve segmentation performance with minimal computational overhead. We validate our approach across multiple 3D semantic segmentation benchmarks, demonstrating consistent performance gains, including up to +11.5% mIoU on TS40K with PTV3, while adding only 58K extra parameters. Our results highlight the benefit of explicitly encoding geometric structure to support accurate and efficient 3D scene understanding, with a lightweight add-on layer
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。