Improved Convex Decomposition with Ensembling and Boolean Primitives

Vaibhav Vavilala, Bodo Rosenhahn, David Forsyth, Florian Kluger, Seemandhar Jain

Keywords: Geometric Primitives, convex decomposition, Ensembling

TL;DR: We present a method to get better primitives with fewer parameters, and introduce ensembling and boolean primitives

Abstract: Describing a scene in terms of primitives -- geometrically simple shapes that offer a parsimonious but accurate abstraction of structure -- is an established and difficult fitting problem. Different scenes require different numbers of primitives, and these primitives interact strongly; however, any proposed solution can be evaluated at inference time. The state of the art method involves a learned regression procedure to predict a start point consisting of a fixed number of primitives, followed by a descent method to refine the geometry and remove redundant primitives. Methods are evaluated by accuracy in depth and normal prediction and in scene segmentation. This paper shows that very significant improvements in accuracy can be obtained by (a) incorporating a small number of \emph{negative} primitives and (b) ensembling over a number of different regression procedures. Ensembling is by refining each predicted start point, then choosing the best by fitting loss. Extensive experiments on the standard NYUv2 dataset confirm that negative primitives are useful, and that our refine-then-choose strategy outperforms choose-then-refine, confirming that the fitting problem is very difficult. Our ensembling with boolean primitives approach strongly outperforms existing methods; additionally we present several improvements to the underlying primitive generation process enabling us to obtain better decompositions with fewer primitives. Code will be released upon acceptance of the paper.

Primary Area: applications to computer vision, audio, language, and other modalities

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 5839