Voxel-based 3D Facies Segmentation from Seismic Data: A Comparative Study

Duc-Thanh Pham; Minh-Tan Pham; Anh Nguyen; Van Nguyen

cs.CVAug 14, 2026

Voxel-based 3D Facies Segmentation from Seismic Data: A Comparative Study

Authors: Duc-Thanh Pham, Minh-Tan Pham, Anh Nguyen, Van Nguyen

Organizations: FPT Software AI Center, Hanoi, Vietnam · Institut de Recherche en Informatique et Syst`emes Al´eatoires (IRISA), UMR 6074, Universit´e Bretagne Sud, 56000 Vannes, France · Department of Computer Science, University of Liverpool, Liverpool, United Kingdom

Abstract

Seismic facies segmentation has emerged as a significant challenge in geophysics, requiring robust methods and systems to effectively identify geologically analogous facies with limited labeled data. Although existing studies have shown promising results in 2D facies segmentation, they often preprocess the original 3D seismic volumes into sets of 2D slices, typically the inline and crossline directions, and treat this problem as a purely 2D segmentation task. This simplification introduces discontinuities across slices and fails to preserve the spatial and structural continuity in 3D seismic data, thus limiting the model's ability to learn coherent geological patterns. In this work, we present a comparative and reproducible benchmark for voxel-based 3D seismic facies segmentation, built upon publicly available seismic volumes including the Netherlands F3 and the Parihaka datasets, with standardized data splits and evaluation metrics. By evaluating the three representative families of modern 3D segmentation architectures, we establish strong baseline results that highlight the potential and remaining challenges for future research in this domain.

Figures & tables

Panel of Figure 1 — Fig. 1 : Distribution of facies classes from the two datasets.

Dataset	Location	Characteristics	Data shape	Fold shape	Train/Val samples
F3	North Sea, Netherlands	Shallow marine, moderate structural variation	601 $\times$ 901 $\times$ 255	601 $\times$ 180 $\times$ 255	975 / 195
Parihaka	Taranaki Basin, New Zealand	High structural complexity: faults, gas hydrates	590 $\times$ 781 $\times$ 1006	590 $\times$ 156 $\times$ 1006	1215 / 405

TABLE I : Summary of the two studied datasets.

TABLE II : Performance of the three methods for voxel-based 3D facies segmentation on two studied datasets. The best results are shown in bold .

Panel of Figure 4 — Fig. 2 : Qualitative comparison of facies segmentation results obtained by the three methods.

Model	Parameters (M)	Train Memory (GB)	GFLOPs
3D Unet	2.51	5.79	95.41
UNETR	146.60	24.92	1868.45
SegMamba	67.36	33.44	1427.20

TABLE III : Comparison of model complexity and computational cost.

Input size	Class 1	Class 2	Class 3	Class 4	Class 5	Class 6	Avg
$96^{3}$	84.39	85.60	86.65	85.54	76.86	82.76	83.63
$128^{3}$	96.67	88.02	96.40	88.09	77.07	79.06	87.69
$144^{3}$	92.89	84.54	96.14	82.67	74.21	79.77	85.04

TABLE IV : Accuracy (%) of 3D U-Net with different training data shapes on the Netherlands F3 dataset using five-fold cross-validation. The best results are shown in bold .

3D Semantic Segmentation

3D semantic segmentation assigns meaning to points, voxels, meshes, or reconstructed scene elements for spatial understanding.

See all 137 papers

UnderOneFacade: Worldwide Facade Semantic Segmentation Benchmark Dataset
Jul 2, 2026Yi Wang, Fan Wang, Prabin Gyawali +93D Point Cloud Segmentation 3D Semantic Segmentation
Technical University of Munich, Munich, Germany · University of Nottingham, Nottingham, UK · CV4DT, University of Cambridge, Cambridge, UK +1
VoxShield: Protecting 3D Medical Datasets from Unauthorized Training via Frequency-Aware Inter-Slice Disruption
May 17, 2026Xinyao Liu, Zhipeng Deng, Wenhan Jiang +4Adversarial Examples 3D Medical Image Segmentation
Westlake University, Hangzhou, China · Dalian University of Technology, Dalian, China · Hokkaido University, Sapporo, Japan +2
SAMamba3D: adapting Segment Anything for generalizable 3D segmentation of multiphase pore-scale images
Apr 29, 2026Rui Zhang, Xianzhi Song, Linqi Zhu +3Image Segmentation 3D Semantic Segmentation
Department of Earth Science and Engineering, Imperial College London, London SW7 2AZ, United Kingdom · College of Petroleum Engineering, China University of Petroleum (Beijing), Beijing 102249, China

Jul 2, 2026J/K move · Enter open · S save

UnderOneFacade: Worldwide Facade Semantic Segmentation Benchmark Dataset

Yi Wang, Fan Wang, Prabin Gyawali, Ziyang Xu+8

Technical University of Munich, Munich, Germany · University of Nottingham, Nottingham, UK · CV4DT, University of Cambridge, Cambridge, UK+1

3D Point Cloud Segmentation 3D Semantic Segmentation Cross-Domain Generalization

Open paper AskPDF arXiv

Globally consistent semantic digital twins require centimeter-accurate and geographically transferable 3D facade segmentation. However, progress in facade parsing is limited by the lack of large-scale, standardized benchmarks for evaluating cross-domain generalization. Existing datasets are geographically narrow, semantically inconsistent, or insufficiently precise. We introduce UnderOneFacade, the largest cross-country and cross-continent 3D facade benchmark to date, comprising centimeter-accurate point clouds with hierarchical, harmonized, and architecturally grounded semantic labels totaling 2.7 billion annotated points. Through a systematic evaluation of representative point-, graph- and transformer-based architectures, we show that current methods struggle to recognize fine-grained architectural elements and degrade significantly across geographic domains, with the best models achieving only up to 33 IoU on the fine-grained LoFG3 benchmark. By combining geometric precision with standardized semantics at unprecedented scale, UnderOneFacade establishes a rigorous benchmark for developing robust and transferable 3D segmentation models. The dataset, evaluation scripts, and pretrained models will be released upon publication.

Abstract

3D Semantic Segmentation

Explore similar work

UnderOneFacade: Worldwide Facade Semantic Segmentation Benchmark Dataset

VoxShield: Protecting 3D Medical Datasets from Unauthorized Training via Frequency-Aware Inter-Slice Disruption

SAMamba3D: adapting Segment Anything for generalizable 3D segmentation of multiphase pore-scale images