Octree Transformer: Autoregressive 3D Shape Generation on Hierarchically Structured Sequences

Moritz Ibing, Gregor Kobsik, Leif Kobbelt; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2023, pp. 2698-2707

Abstract


Autoregressive models have proven to be very powerful in NLP text generation tasks and lately have gained popularity for image generation as well. However, they have seen limited use for the synthesis of 3D shapes so far. This is mainly due to the lack of a straightforward way to linearize 3D data as well as to scaling problems with the length of the resulting sequences when describing complex shapes. In this work we address both of these problems. We use octrees as a compact hierarchical shape representation that can be sequentialized by traversal ordering. Moreover, we introduce an adaptive compression scheme, that significantly reduces sequence lengths and thus enables their effective generation with a transformer, while still allowing fully autoregressive sampling and parallel training. We demonstrate the performance of our model by comparing against the state-of-the-art in shape generation.

Related Material


[pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Ibing_2023_CVPR, author = {Ibing, Moritz and Kobsik, Gregor and Kobbelt, Leif}, title = {Octree Transformer: Autoregressive 3D Shape Generation on Hierarchically Structured Sequences}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {2698-2707} }