Neural Style Transfer for Point Clouds
This work addresses style transfer for point clouds, enabling editing of 3D data, but it is incremental as it adapts existing image-based techniques to point clouds.
The authors tackled the problem of editing geometric or color properties of point clouds by proposing a neural style transfer method that transfers style from one point cloud or image to another, using a pre-trained PointNet-based network and Gram-based representations, with experimental results showing capability for single objects and indoor scenes.
How can we edit or transform the geometric or color property of a point cloud? In this study, we propose a neural style transfer method for point clouds which allows us to transfer the style of geometry or color from one point cloud either independently or simultaneously to another. This transfer is achieved by manipulating the content representations and Gram-based style representations extracted from a pre-trained PointNet-based classification network for colored point clouds. As Gram-based style representation is invariant to the number or the order of points, the same method can be extended to transfer the style extracted from an image to the color expression of a point cloud by merely treating the image as a set of pixels. Experimental results demonstrate the capability of the proposed method for transferring style from either an image or a point cloud to another point cloud of a single object or even an indoor scene.