Symbol as Points: Panoptic Symbol Spotting via Point-based Representation

TL;DR

This paper introduces SymPoint, a point-based method for panoptic symbol spotting in CAD drawings, utilizing point transformers and novel modules to improve accuracy over existing image-based and graph-based approaches.

Contribution

It proposes a novel point cloud segmentation approach with attention and contrastive learning modules for symbol spotting in CAD drawings, outperforming state-of-the-art methods.

Findings

Achieves 9.6% higher PQ than GAT-CADNet on FloorPlanCAD dataset.

Effectively handles primitive mask downsampling with KNN interpolation.

Introduces attention with connection and contrastive connection learning modules.

Abstract

This work studies the problem of panoptic symbol spotting, which is to spot and parse both countable object instances (windows, doors, tables, etc.) and uncountable stuff (wall, railing, etc.) from computer-aided design (CAD) drawings. Existing methods typically involve either rasterizing the vector graphics into images and using image-based methods for symbol spotting, or directly building graphs and using graph neural networks for symbol recognition. In this paper, we take a different approach, which treats graphic primitives as a set of 2D points that are locally connected and use point cloud segmentation methods to tackle it. Specifically, we utilize a point transformer to extract the primitive features and append a mask2former-like spotting head to predict the final output. To better use the local connection information of primitives and enhance their discriminability, we further propose the attention with connection module (ACM) and contrastive connection learning scheme (CCL). Finally, we propose a KNN interpolation mechanism for the mask attention module of the spotting head to better handle primitive mask downsampling, which is primitive-level in contrast to pixel-level for the image. Our approach, named SymPoint, is simple yet effective, outperforming recent state-of-the-art method GAT-CADNet by an absolute increase of 9.6% PQ and 10.4% RQ on the FloorPlanCAD dataset. The source code and models will be available at https://github.com/nicehuster/SymPoint.