Position: You Can’t Manufacture a NeRF

Ma Kimmel, Mueed Ur Rehman, Yonatan Bisk, Gary K. Fedder
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:81652-81664, 2025.

Abstract

In this paper, we examine the manufacturability gap in state-of-the-art generative models for 3D object representations. Many models for generating 3D assets focus on rendering virtual content and do not consider the constraints of real-world manufacturing, such as milling, casting, or injection molding. We demonstrate that existing generative models for computer-aided design representation do not generalize outside of their training datasets or to unmodified real, human-created objects. We identify limitations with the current approaches, including missing manufacturing-readable semantics, the inability to decompose complex shapes into parameterized segments appropriate for computer-aided manufacturing, and a lack of appropriate scoring metrics to assess the generated output versus the true reconstruction. The academic community could greatly impact real-world manufacturing by rallying around pathways to solve these challenges. We offer revised, more realistic datasets and baseline benchmarks as a step in targeting the challenge. In evaluating these datasets, we find that existing models are severely overfit to simpler data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-kimmel25a, title = {Position: You Can’t Manufacture a {N}e{RF}}, author = {Kimmel, Ma and Rehman, Mueed Ur and Bisk, Yonatan and Fedder, Gary K.}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {81652--81664}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/kimmel25a/kimmel25a.pdf}, url = {https://proceedings.mlr.press/v267/kimmel25a.html}, abstract = {In this paper, we examine the manufacturability gap in state-of-the-art generative models for 3D object representations. Many models for generating 3D assets focus on rendering virtual content and do not consider the constraints of real-world manufacturing, such as milling, casting, or injection molding. We demonstrate that existing generative models for computer-aided design representation do not generalize outside of their training datasets or to unmodified real, human-created objects. We identify limitations with the current approaches, including missing manufacturing-readable semantics, the inability to decompose complex shapes into parameterized segments appropriate for computer-aided manufacturing, and a lack of appropriate scoring metrics to assess the generated output versus the true reconstruction. The academic community could greatly impact real-world manufacturing by rallying around pathways to solve these challenges. We offer revised, more realistic datasets and baseline benchmarks as a step in targeting the challenge. In evaluating these datasets, we find that existing models are severely overfit to simpler data.} }
Endnote
%0 Conference Paper %T Position: You Can’t Manufacture a NeRF %A Ma Kimmel %A Mueed Ur Rehman %A Yonatan Bisk %A Gary K. Fedder %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-kimmel25a %I PMLR %P 81652--81664 %U https://proceedings.mlr.press/v267/kimmel25a.html %V 267 %X In this paper, we examine the manufacturability gap in state-of-the-art generative models for 3D object representations. Many models for generating 3D assets focus on rendering virtual content and do not consider the constraints of real-world manufacturing, such as milling, casting, or injection molding. We demonstrate that existing generative models for computer-aided design representation do not generalize outside of their training datasets or to unmodified real, human-created objects. We identify limitations with the current approaches, including missing manufacturing-readable semantics, the inability to decompose complex shapes into parameterized segments appropriate for computer-aided manufacturing, and a lack of appropriate scoring metrics to assess the generated output versus the true reconstruction. The academic community could greatly impact real-world manufacturing by rallying around pathways to solve these challenges. We offer revised, more realistic datasets and baseline benchmarks as a step in targeting the challenge. In evaluating these datasets, we find that existing models are severely overfit to simpler data.
APA
Kimmel, M., Rehman, M.U., Bisk, Y. & Fedder, G.K.. (2025). Position: You Can’t Manufacture a NeRF. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:81652-81664 Available from https://proceedings.mlr.press/v267/kimmel25a.html.

Related Material