TempFit: Panoptic Symbol Spotting via Template Fitting
Open Access
- 9 November 2026
- journal article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (6) , 1-14
Abstract
In this paper, we introduce TempFit , a novel approach for the problem of Panoptic Symbol Spotting on CAD drawings. Existing methods treat symbol spotting as a per-primitive classification or mask-prediction task, with no explicit constraint enforcing that the primitives assigned to one instance form a spatially connected group. As a result, predictions may erroneously merge primitives from distant symbols into a single instance or fragment a coherent symbol across multiple predictions. Motivated by the professional "retrieve-and-transform" CAD design workflow, we reformulate this task as a Template Fitting problem, where each foreground instance is detected by retrieving and aligning a canonical template from a library, enforcing structural completeness by construction. Building on this formulation, we present an autoregressive transformer that unifies template retrieval and pose prediction within a single end-to-end pipeline. The framework treats the template library as an explicit geometric memory, parsing background regions in parallel via learnable background tokens and synthesizing foreground instances autoregressively while attending to template-specific geometry at each step. Experiments on FloorplanCAD and our newly collected IndustryCAD dataset, a more challenging benchmark with complex industrial symbols, heavy occlusion, and wide scale variation, demonstrate state-of-the-art performance with particularly large gains on IndustryCAD. Project page: https://tempfit-project.github.io/.Keywords
Funding Information
- National Key R&D Program of China (2024YFB2809102)
- NSFC (62322207)
- NSFC (62325211)
- NSFC (62132021)
- Research Fund of Jiangsu Key Laboratory of AI for Industries (E6420016G8)
- Lakehead University (Faculty of Science & Environmental Studies)
This publication has 29 references indexed in Scilit:
- Shape embedding and retrieval in multi-flow deformationComputational Visual Media, 2024
- A survey of deep learning-based 3D shape generationComputational Visual Media, 2023
- ParSeNet: A Parametric Surface Fitting Network for 3D Point Clouds2020
- End-to-End Object Detection with Transformers2020
- Semi-Supervised Learning With Graph Learning-Convolutional Networks2019
- Scan2CAD: Learning CAD Model Alignment in RGB-D Scans2019
- Learning Shape Abstractions by Assembling Volumetric Primitives2017
- Generation of synthetic documents for performance evaluation of symbol recognition & spotting systemsInternational Journal on Document Analysis and Recognition (IJDAR), 2010
- A method for registration of 3-D shapesIEEE Transactions on Pattern Analysis and Machine Intelligence, 1992
- Morphological shape decompositionIEEE Transactions on Pattern Analysis and Machine Intelligence, 1990