We introduce SketchGNN , a convolutional graph neural network for semantic segmentation and labeling of freehand vector sketches. We treat an input stroke-based sketch as a graph with nodes representing the sampled points along input strokes and edges encoding the stroke structure information. To predict the per-node labels, our SketchGNN uses graph convolution and a static-dynamic branching network architecture to extract the features at three levels, i.e., point-level, stroke-level, and sketch-level. SketchGNN significantly improves the accuracy of the state-of-the-art methods for semantic sketch segmentation (by 11.2% in the pixel-based metric and 18.2% in the component-based metric over a large-scale challenging SPG dataset) and has magnitudes fewer parameters than both image-based and sequence-based methods.

We introduce SketchGCN, a graph convolutional neural network for semantic segmentation and labeling of freehand sketches. We treat an input sketch as a 2D point set, and encode the stroke structure information into graph node/edge representations. To predict the per-point labels, our SketchGCN uses graph convolution and a globallocal branching network architecture to extract both intrastroke and inter-stroke features. SketchGCN significantly improves the accuracy of the state-of-the-art methods for semantic sketch segmentation (by 11.4% in the pixel-based metric and 18.2% in the component-based metric over a large-scale challenging SPG dataset) and has magnitudes fewer parameters than both image-based and sequencebased methods.

The diffusion coatings were deposited on commercially pure Ti and Ti-6Al-4V alloy at up to 1000 °C for up to 10 h using the pack cementation method. The pack powders consisted of 4 wt% Al (Al reservoir) and 4 wt% NH4Cl (activator) which were balanced with Al2O3 (inert filler). The growth kinetics of coatings were gravimetrically measured by a high precision balance. The aluminised specimens were characterised by means of scanning electron microscopy (SEM), energy dispersive spectrometer (EDS) and X-ray diffraction (XRD). At the early stages of deposition, a TiO2 (rutile) scale, other than aluminide coating, was developed on both materials at <900 °C. As the experimental temperature arose above 900 °C, the rutile layer became unstable and reduced to the low oxidation state of Ti oxides. When the temperature increased to 1000 °C, the TiO2 scale dissociated almost completely and the aluminide coating began to develop. After a triple-layered coating was generated, the coating growth was governed by the outward migration of Ti species from the substrates and obeyed the parabolic law. The coating formed consisted of an outer layer of Al3Ti, a mid-layer of Al2Ti and an inner layer of AlTi. The outer layer of Al3Ti dominated the thickness of the aluminide coating.