Trained a watermark recognition model specifically for vehicle images on the DCD Automotive website, which includes the words' DCD '.
The underlying framework is generated based on the YoLoV8 visual recognition framework, and the dataset is manually annotated with 350 image watermarks.
