Much of the attention surrounding deep learning and its impact on visual effects has focused on image-based techniques, in particular the compelling examples produced by denoising, segmentation, and generative adversarial models. Research continues, however, into application of machine learning to 3D data – meshes, point clouds, voxel grids, and so on – and how to leverage functional components which have demonstrated their usefulness on images. Appropriate data representation remains a challenge, but here is a great overview of the progress being made.
3D data is a valuable asset in the field of computer vision as it provides
rich information about the full geometry of sensed objects and scenes. With the recent availability of large 3D datasets and the increase in computational power, it is today possible to consider applying deep learning to learn specific tasks on 3D data such as segmentation, recognition and correspondence.
Depending on the considered 3D data representation, different challenges may be foreseen in using existent deep learning architectures. In this paper, we provide a comprehensive overview of various 3D data representations
highlighting the difference between Euclidean and non-Euclidean ones. We also discuss how deep learning methods are applied on each representation, analyzing the challenges to overcome.
Have you seen some amazing research which should be covered here? Contact us or let us know in the comments section below!