DrivAerNet++: Sketches

In addition to the 3D mesh data, our dataset includes diverse sketch representations of automotive designs through two complementary approaches: Canny edge detection and CLIPasso-generated hand-drawn sketches. The Canny edge sketches provide precise, consistent edge-based representations that capture the fundamental geometric features and contours of each vehicle design. The CLIPasso sketches offer more organic, hand-drawn style representations that mimic human artistic interpretation of the automotive forms. These dual sketch modalities are instrumental for a range of machine learning tasks, including sketch-to-3D generation, cross-modal retrieval, style transfer, and multi-modal learning applications. The comprehensive sketch representations can also facilitate automated design workflows by providing both technical edge information and artistic interpretation of vehicle geometries. By incorporating these complementary sketch styles, our dataset enhances the utility for developing and testing advanced algorithms in automotive design, computer vision, and human-computer interaction for engineering applications. Strict Licensing Notice: DrivAerNet/DrivAerNet++ is released under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0) and is exclusively for non-commercial research and educational purposes. Any commercial use—including, but not limited to, training machine learning models, developing generative AI tools, creating software products, running new simulations using the provided geometries or any derived geometries, or other commercial R&D applications—is strictly prohibited. Unauthorized commercial use of DrivAerNet/DrivAerNet++, or any derived data, will result in enforcement by the MIT Technology Licensing Office (MIT TLO) and may carry legal consequences.