iRBSM: A Deep Implicit 3D Breast Shape Model

TL;DR

The paper introduces iRBSM, a deep implicit neural model for 3D female breast shapes that improves upon previous PCA-based models by capturing detailed geometry without complex registration, enabling single-image reconstruction.

Contribution

It presents the first deep implicit 3D breast shape model, eliminating the need for non-rigid registration and enhancing surface detail capture compared to prior PCA-based models.

Findings

Outperforms RBSM in surface reconstruction tasks.

Captures detailed surface features like nipples and belly buttons.

Enables 3D breast shape reconstruction from a single image.

Abstract

We present the first deep implicit 3D shape model of the female breast, building upon and improving the recently proposed Regensburg Breast Shape Model (RBSM). Compared to its PCA-based predecessor, our model employs implicit neural representations; hence, it can be trained on raw 3D breast scans and eliminates the need for computationally demanding non-rigid registration -- a task that is particularly difficult for feature-less breast shapes. The resulting model, dubbed iRBSM, captures detailed surface geometry including fine structures such as nipples and belly buttons, is highly expressive, and outperforms the RBSM on different surface reconstruction tasks. Finally, leveraging the iRBSM, we present a prototype application to 3D reconstruct breast shapes from just a single image. Model and code publicly available at https://rbsm.re-mic.de/implicit.