BIKED++: A Multimodal Dataset of 1.4 Million Bicycle Image and Parametric CAD Designs

TL;DR

This paper presents BIKED++, a large multimodal dataset of 1.4 million bicycle designs in parametric, image, and text formats, enabling cross-modal learning and similarity estimation between designs and descriptions.

Contribution

The paper introduces a comprehensive multimodal bicycle dataset with a rendering engine and trained models for cross-modal prediction, advancing design analysis and retrieval.

Findings

Successfully trained models to estimate CLIP embeddings from parametric designs.

Enabled similarity comparisons between bicycle designs and textual or visual references.

Provided publicly available dataset and tools for multimodal bicycle design research.

Abstract

This paper introduces a public dataset of 1.4 million procedurally-generated bicycle designs represented parametrically, as JSON files, and as rasterized images. The dataset is created through the use of a rendering engine which harnesses the BikeCAD software to generate vector graphics from parametric designs. This rendering engine is discussed in the paper and also released publicly alongside the dataset. Though this dataset has numerous applications, a principal motivation is the need to train cross-modal predictive models between parametric and image-based design representations. For example, we demonstrate that a predictive model can be trained to accurately estimate Contrastive Language-Image Pretraining (CLIP) embeddings from a parametric representation directly. This allows similarity relations to be established between parametric bicycle designs and text strings or reference images. Trained predictive models are also made public. The dataset joins the BIKED dataset family which includes thousands of mixed-representation human-designed bicycle models and several datasets quantifying design performance. The code and dataset can be found at: https://github.com/Lyleregenwetter/BIKED_multimodal/tree/main