GP22: A Car Styling Dataset for Automotive Designers
This addresses the problem of inefficient design data archiving for automotive designers, though it is incremental as it adapts existing methods to a new domain-specific dataset.
The authors tackled the lack of automotive design-specific datasets by releasing GP22, a dataset of 1480 car side profile images with annotations defined by designers, and achieved a baseline model with an mAP of 0.995 and recall of 0.984.
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.