Deep Learning-Meets-Active Surfaces: A Unified Framework for Multi-Object Segmentation and Shape Analysis in 3D Fluorescence Microscopy
Abstract
In this paper, we introduce Nagini3D, a 3D biological image segmentation method that hybridizes two high-performing yet highly complementary segmentation paradigms: deep learning-based approaches and snake-based methods. Our approach leverages both the high efficiency of data-driven techniques and the convenient shape representation provided by active surfaces, which proves highly practical for incorporating a priori knowledge as well as for the post-analysis of segmented objects. By considering a complete 3D volume, our method ensures segmentation consistency across all three dimensions, unlike many approaches that handle volumes in a slice-by-slice manner. Additionally, it theoretically enables the representation of any object with spherical topology, overcoming the limitations of common 3D methods that rely on object convexity assumptions. Furthermore, the continuous representation of objects allows for a resolution-independent description, facilitating the extraction of various geometric features of interest (e.g., local curvature, principal deformations). We demonstrate both the quality of the segmentation produced by our method and the richness of the accessible geometric features through an extensive set of experiments conducted on multiple real and simulated datasets, including objects with challenging geometries.
Related
- Automatic Generation of Expert-Level Neuron Segmentation Masks from Fluorescence Microscopy Images for Non-Invasive Deep Learning Analysis of Phase-Contrast Images
- Supervised 3D Virtual Fluorescence and Pseudo-H&E Staining of Fresh Bulk Tissues via Multimodal qOBM and Fluorescence Microscopy
- Quanta microscopy enables fluorescence imaging at the native bit-plane rate
- Live-Imaging of Molecular Processes with Real-Time Fluorescence Microscopy
- Visualizing Dolichogenidea gelechiidivoris parasitism within Tuta absoluta larvae through fluorescence microscopy in optically cleared samples
BibTeX
@misc{rapilly2026deep,
title = {{Deep Learning-Meets-Active Surfaces: A Unified Framework for Multi-Object Segmentation and Shape Analysis in 3D Fluorescence Microscopy}},
author = {Quentin Rapilly and Anaïs Badoual and You-Hsuan Liu and Charles Kervrann},
howpublished = {HAL},
year = {2026},
month = sep,
doi = {10.64898/2026.09.30.755627},
url = {https://hal.science/hal-05784325v1},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Deep Learning-Meets-Active Surfaces: A Unified Framework for Multi-Object Segmentation and Shape Analysis in 3D Fluorescence Microscopy
AU - Quentin Rapilly
AU - Anaïs Badoual
AU - You-Hsuan Liu
AU - Charles Kervrann
PY - 2026
DA - 2026/09/30
PB - HAL
DO - 10.64898/2026.09.30.755627
UR - https://hal.science/hal-05784325v1
AB - In this paper, we introduce Nagini3D, a 3D biological image segmentation method that hybridizes two high-performing yet highly complementary segmentation paradigms: deep learning-based approaches and snake-based methods. Our approach leverages both the high efficiency of data-driven techniques and the convenient shape representation provided by active surfaces, which proves highly practical for incorporating a priori knowledge as well as for the post-analysis of segmented objects. By considering a complete 3D volume, our method ensures segmentation consistency across all three dimensions, unlike many approaches that handle volumes in a slice-by-slice manner. Additionally, it theoretically enables the representation of any object with spherical topology, overcoming the limitations of common 3D methods that rely on object convexity assumptions. Furthermore, the continuous representation of objects allows for a resolution-independent description, facilitating the extraction of various geometric features of interest (e.g., local curvature, principal deformations). We demonstrate both the quality of the segmentation produced by our method and the richness of the accessible geometric features through an extensive set of experiments conducted on multiple real and simulated datasets, including objects with challenging geometries.
ER -