Stitching and dimensionality effects on large artificially generated volume datasets

• URL: http://arxiv.org/abs/2606.20095v1
• Date: Thu, 18 Jun 2026 11:14:20 GMT
• Title: Stitching and dimensionality effects on large artificially generated volume datasets
• Abstract summary: stitching artifacts can affect generative models for style-transfer.<n>We investigated three stitching approaches and two patch dimensionalities (2D vs 3D) using cycleGAN models trained on cryo-electron microscopy datasets.<n>We evaluated both perceptual quality and performance on downstream mitochondria segmentation.
• Score: 0.9664596800095625
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Generating large images via deep learning requires patching input data to accommodate hardware memory limitations, then assembling output patches, a process that can introduce stitching artifacts when neighboring patches do not align at borders. While these artifacts are known to affect segmentation tasks, their impact on generative models for style-transfer remains poorly understood. We investigated three stitching approaches and two patch dimensionalities (2D vs 3D) using cycleGAN models trained on cryo-electron microscopy datasets. We evaluated both perceptual quality and performance on downstream mitochondria segmentation. Our key findings reveal that: (1) FID scores fail to detect subtle stitching artifacts that significantly impact downstream segmentation performance, (2) 3D models with artifact-free stitching marginally outperform 2D models on downstream tasks, though the improvement barely justifies the computational cost, and (3) 2D models train more stably due to larger batch sizes. Additionally, we demonstrate that ensembling predictions from three orthogonal directions can improve low-quality volumes but provides no benefit for high-quality outputs. These results demonstrate that maximizing generative model performance on large scientific datasets requires careful consideration and mitigation of stitching artifacts, and that perceptual metrics alone are insufficient for evaluating domain adaptation quality in biomedical imaging.

Related papers

• Damage Classification for 3D Point Cloud Data via 3D Data Analysis and Vision Foundation Model-based 2D Projections [4.959316347509111]
  This study examines two methods: 3D PCD-based damage assessment (3PDA) algorithm and 2D projection damage assessment (2PDA)<n>In our 3PDA analysis algorithm, TDA is used to derive compact representations of 3D PCD segmented by pointNet.<n>In the 2D projection analysis algorithm, we leverage large VFMs for granular damage detection by projecting 3D PCD into 2D views.
  arXiv  Detail & Related papers  (2026-08-09T23:18:42Z)
• Perceptual Quality Assessment of 3D Gaussian Splatting: A Subjective Dataset and Prediction Metric [76.66966098297986]
  We present 3DGS-QA, the first subjective quality assessment dataset for 3DGS.<n>It comprises 225 degraded reconstructions across 15 object types, enabling a controlled investigation of common distortion factors.<n>Our model extracts spatial and photometric cues from the Gaussian representation to estimate perceived quality in a structure-aware manner.
  arXiv  Detail & Related papers  (2025-11-11T09:34:20Z)
• Print2Volume: Generating Synthetic OCT-based 3D Fingerprint Volume from 2D Fingerprint Image [5.697372349288325]
  This paper introduces Print2Volume, a novel framework for generating realistic, synthetic 3D fingerprints from 2D fingerprint image.<n>We generated a large-scale synthetic dataset of 420,000 samples.<n>By pre-training a recognition model on our synthetic data and fine-tuning it on a small real-world dataset, we achieved a remarkable reduction in the Equal Error Rate (EER)
  arXiv  Detail & Related papers  (2025-08-29T07:26:39Z)
• Plug-and-Play with 2.5D Artifact Reduction Prior for Fast and Accurate Industrial Computed Tomography Reconstruction [4.024850952459758]
  Cone-beam X-ray computed tomography (XCT) is an essential imaging technique for generating 3D reconstructions of internal structures.<n>Recent work incorporating artifact reduction priors within a plug-and-play tomography reconstruction framework has shown promising results.<n>We show that this 2.5D prior not only improves the quality of reconstructions but also enables the model to directly suppress commonly occurring XCT artifacts.
  arXiv  Detail & Related papers  (2025-06-17T16:52:57Z)
• P3Net: Progressive and Periodic Perturbation for Semi-Supervised Medical Image Segmentation [60.08541107831459]
  We propose a progressive and periodic perturbation mechanism (P3M) and a boundary-focused loss to guide the learning of unlabeled data.<n>Our method achieves state-of-the-art performance on two 2D and 3D datasets.
  arXiv  Detail & Related papers  (2025-05-21T05:35:28Z)
• Ray Denoising: Depth-aware Hard Negative Sampling for Multi-view 3D Object Detection [46.041193889845474]
  Ray Denoising is an innovative method that enhances detection accuracy by strategically sampling along camera rays to construct hard negative examples. Ray Denoising is designed as a plug-and-play module, compatible with any DETR-style multi-view 3D detectors. It achieves a 1.9% improvement in mean Average Precision (mAP) over the state-of-the-art StreamPETR method on the NuScenes dataset.
  arXiv  Detail & Related papers  (2024-02-06T02:17:44Z)
• FILP-3D: Enhancing 3D Few-shot Class-incremental Learning with Pre-trained Vision-Language Models [59.13757801286343]
  Few-shot class-incremental learning aims to mitigate the catastrophic forgetting issue when a model is incrementally trained on limited data.<n>We introduce the FILP-3D framework with two novel components: the Redundant Feature Eliminator (RFE) for feature space misalignment and the Spatial Noise Compensator (SNC) for significant noise.
  arXiv  Detail & Related papers  (2023-12-28T14:52:07Z)
• Interpretable 2D Vision Models for 3D Medical Images [47.75089895500738]
  This study proposes a simple approach of adapting 2D networks with an intermediate feature representation for processing 3D images. We show on all 3D MedMNIST datasets as benchmark and two real-world datasets consisting of several hundred high-resolution CT or MRI scans that our approach performs on par with existing methods.
  arXiv  Detail & Related papers  (2023-07-13T08:27:09Z)
• Voxel-wise classification for porosity investigation of additive manufactured parts with 3D unsupervised and (deeply) supervised neural networks [5.467497693327066]
  This study revisits recent supervised (UNet, UNet++, UNet 3+, MSS-UNet) and unsupervised (VAE, ceVAE, gmVAE, vqVAE) DL models for volumetric analysis of AM samples from X-CT images. It extends them to accept 3D input data with a 3D-patch pipeline for lower computational requirements, improved efficiency and generalisability. The VAE/ceVAE models demonstrated superior capabilities, particularly when leveraging post-processing techniques.
  arXiv  Detail & Related papers  (2023-05-13T11:23:00Z)
• Towards Model Generalization for Monocular 3D Object Detection [57.25828870799331]
  We present an effective unified camera-generalized paradigm (CGP) for Mono3D object detection. We also propose the 2D-3D geometry-consistent object scaling strategy (GCOS) to bridge the gap via an instance-level augment. Our method called DGMono3D achieves remarkable performance on all evaluated datasets and surpasses the SoTA unsupervised domain adaptation scheme.
  arXiv  Detail & Related papers  (2022-05-23T23:05:07Z)
• RiCS: A 2D Self-Occlusion Map for Harmonizing Volumetric Objects [68.85305626324694]
  Ray-marching in Camera Space (RiCS) is a new method to represent the self-occlusions of foreground objects in 3D into a 2D self-occlusion map. We show that our representation map not only allows us to enhance the image quality but also to model temporally coherent complex shadow effects.
  arXiv  Detail & Related papers  (2022-05-14T05:35:35Z)
• Secrets of 3D Implicit Object Shape Reconstruction in the Wild [92.5554695397653]
  Reconstructing high-fidelity 3D objects from sparse, partial observation is crucial for various applications in computer vision, robotics, and graphics. Recent neural implicit modeling methods show promising results on synthetic or dense datasets. But, they perform poorly on real-world data that is sparse and noisy. This paper analyzes the root cause of such deficient performance of a popular neural implicit model.
  arXiv  Detail & Related papers  (2021-01-18T03:24:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.