Position: You Can't Manufacture a NeRF

MA Kimmel, Mueed Ur Rehman, Yonatan Bisk, Gary K. Fedder

TL;DR: Complex AI-Generated 3D objects are currently not in mass-manufacturable form (and often not even 3D printable without heavy editing), and simple generated CAD objects aren't realistic -- we should fix this.

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.

Lay Summary: Most reconstructive and generative AI for 3D objects targets creating objects that are only fit for being rendered on the screen, they do not create files in a format that can be created in the real world via common, modern manufacturing techniques. The ones that do create files in that format (called boundary representation) can only create very simple objects. We demonstrate the limits of existing models in this paper.

Verify Author Names: My co-authors have confirmed that their names are spelled correctly both on OpenReview and in the camera-ready PDF. (If needed, please update ‘Preferred Name’ in OpenReview to match the PDF.)

No Additional Revisions: I understand that after the May 29 deadline, the camera-ready submission cannot be revised before the conference. I have verified with all authors that they approve of this version.

Pdf Appendices: My camera-ready PDF file contains both the main text (not exceeding the page limits) and all appendices that I wish to include. I understand that any other supplementary material (e.g., separate files previously uploaded to OpenReview) will not be visible in the PMLR proceedings.

Latest Style File: I have compiled the camera ready paper with the latest ICML2025 style files <https://media.icml.cc/Conferences/ICML2025/Styles/icml2025.zip> and the compiled PDF includes an unnumbered Impact Statement section.

Paper Verification Code: MmE3Y

Primary Area: Research Priorities, Methodology, and Evaluation

Keywords: Generative AI, 3D Mesh Generation, CAD Reconstruction

Submission Number: 409