Improving Image Tracing with Convolutional Autoencoders by High-Pass Filter Preprocessing
This work addresses image tracing for computer graphics or design applications, but it appears incremental as it combines existing methods like high-pass filtering and autoencoders.
The study tackled image tracing by using high-pass filter preprocessing with convolutional autoencoders to improve vectorization, resulting in increased effectiveness in representing images more abstractly.
The process of transforming a raster image into a vector representation is known as image tracing. This study looks into several processing methods that include high-pass filtering, autoencoding, and vectorization to extract an abstract representation of an image. According to the findings, rebuilding an image with autoencoders, high-pass filtering it, and then vectorizing it can represent the image more abstractly while increasing the effectiveness of the vectorization process.