GP22: A Car Styling Dataset for Automotive Designers

TL;DR

The paper introduces GP22, a specialized dataset of car styling features designed for automotive designers, enabling automated design data archiving and analysis with high accuracy models.

Contribution

It provides a novel dataset focused on car exterior design features with annotations aligned to designer taxonomy, and demonstrates a high-performance detection baseline.

Findings

YOLO v5 achieved an mAP of 0.995

Recall rate of 0.984 on the dataset

Model performance scalable to sketches and renderings

Abstract

An automated design data archiving could reduce the time wasted by designers from working creatively and effectively. Though many datasets on classifying, detecting, and instance segmenting on car exterior exist, these large datasets are not relevant for design practices as the primary purpose lies in autonomous driving or vehicle verification. Therefore, we release GP22, composed of car styling features defined by automotive designers. The dataset contains 1480 car side profile images from 37 brands and ten car segments. It also contains annotations of design features that follow the taxonomy of the car exterior design features defined in the eye of the automotive designer. We trained the baseline model using YOLO v5 as the design feature detection model with the dataset. The presented model resulted in an mAP score of 0.995 and a recall of 0.984. Furthermore, exploration of the model performance on sketches and rendering images of the car side profile implies the scalability of the dataset for design purposes.