Sketch semantic segmentation serves as an important part of sketch interpretation. Recently, some researchers have obtained significant results using graph neural networks (GNN) for this task. However ...

As shown in the comparison given in Figure 11, the model that combines Graph Neural Networks (GNN) with an Encoder-Decoder architecture outperforms the other models, achieving an accuracy of 51.46%.

For auto-encoder, we use the model based on multiple MLP encoder layers and multiple MLP decoder layers, introduced by Hinton [21]. In the encoder part, each layer’s size is reduced by half compared ...