Rep3D: Re-parameterize Large 3D Kernels with Low-Rank Receptive Modeling for Medical Imaging

• URL: http://arxiv.org/abs/2505.19603v1
• Date: Mon, 26 May 2025 07:12:56 GMT
• Title: Rep3D: Re-parameterize Large 3D Kernels with Low-Rank Receptive Modeling for Medical Imaging
• Authors: Ho Hin Lee, Quan Liu, Shunxing Bao, Yuankai Huo, Bennett A. Landman,
• Abstract summary: Rep3D is a 3D convolutional framework that incorporates a learnable spatial volumetric prior into large kernel training.<n>Rep3D offers an interpretable, and scalable solution for 3D medical image analysis.
• Score: 15.142146104837005
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In contrast to vision transformers, which model long-range dependencies through global self-attention, large kernel convolutions provide a more efficient and scalable alternative, particularly in high-resolution 3D volumetric settings. However, naively increasing kernel size often leads to optimization instability and degradation in performance. Motivated by the spatial bias observed in effective receptive fields (ERFs), we hypothesize that different kernel elements converge at variable rates during training. To support this, we derive a theoretical connection between element-wise gradients and first-order optimization, showing that structurally re-parameterized convolution blocks inherently induce spatially varying learning rates. Building on this insight, we introduce Rep3D, a 3D convolutional framework that incorporates a learnable spatial prior into large kernel training. A lightweight two-stage modulation network generates a receptive-biased scaling mask, adaptively re-weighting kernel updates and enabling local-to-global convergence behavior. Rep3D adopts a plain encoder design with large depthwise convolutions, avoiding the architectural complexity of multi-branch compositions. We evaluate Rep3D on five challenging 3D segmentation benchmarks and demonstrate consistent improvements over state-of-the-art baselines, including transformer-based and fixed-prior re-parameterization methods. By unifying spatial inductive bias with optimization-aware learning, Rep3D offers an interpretable, and scalable solution for 3D medical image analysis. The source code is publicly available at https://github.com/leeh43/Rep3D.

Related papers

• H3R: Hybrid Multi-view Correspondence for Generalizable 3D Reconstruction [39.22287224290769]
  H3R is a hybrid framework that integrates latent fusion with attention-based feature aggregation.<n>By integrating both paradigms, our approach enhances generalization while converging 2$times$ faster than existing methods.<n>Our method supports variable-number and high-resolution input views while demonstrating robust cross-dataset generalization.
  arXiv  Detail & Related papers  (2025-08-05T05:56:30Z)
• Dynamic 3D KAN Convolution with Adaptive Grid Optimization for Hyperspectral Image Classification [12.168520751389622]
  KANet is an improved 3D-DenseNet model, consisting of 3D KAN Conv and an adaptive grid update mechanism.<n> KANet enhances model representation capability through a 3D dynamic expert convolution system without increasing network depth or width.<n>The proposed method demonstrates superior performance on IN, UP, and KSC datasets, outperforming mainstream hyperspectral image classification approaches.
  arXiv  Detail & Related papers  (2025-04-21T14:57:48Z)
• 3D Equivariant Pose Regression via Direct Wigner-D Harmonics Prediction [50.07071392673984]
  Existing methods learn 3D rotations parametrized in the spatial domain using angles or quaternions. We propose a frequency-domain approach that directly predicts Wigner-D coefficients for 3D rotation regression. Our method achieves state-of-the-art results on benchmarks such as ModelNet10-SO(3) and PASCAL3D+.
  arXiv  Detail & Related papers  (2024-11-01T12:50:38Z)
• PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting [54.7468067660037]
  PF3plat sets a new state-of-the-art across all benchmarks, supported by comprehensive ablation studies validating our design choices.<n>Our framework capitalizes on fast speed, scalability, and high-quality 3D reconstruction and view synthesis capabilities of 3DGS.
  arXiv  Detail & Related papers  (2024-10-29T15:28:15Z)
• Graph and Skipped Transformer: Exploiting Spatial and Temporal Modeling Capacities for Efficient 3D Human Pose Estimation [36.93661496405653]
  We take a global approach to exploit Transformer-temporal information with a concise Graph and Skipped Transformer architecture. Specifically, in 3D pose stage, coarse-grained body parts are deployed to construct a fully data-driven adaptive model. Experiments are conducted on Human3.6M, MPI-INF-3DHP and Human-Eva benchmarks.
  arXiv  Detail & Related papers  (2024-07-03T10:42:09Z)
• NDC-Scene: Boost Monocular 3D Semantic Scene Completion in Normalized Device Coordinates Space [77.6067460464962]
  Monocular 3D Semantic Scene Completion (SSC) has garnered significant attention in recent years due to its potential to predict complex semantics and geometry shapes from a single image, requiring no 3D inputs. We identify several critical issues in current state-of-the-art methods, including the Feature Ambiguity of projected 2D features in the ray to the 3D space, the Pose Ambiguity of the 3D convolution, and the Imbalance in the 3D convolution across different depth levels. We devise a novel Normalized Device Coordinates scene completion network (NDC-Scene) that directly extends the 2
  arXiv  Detail & Related papers  (2023-09-26T02:09:52Z)
• Fast-SNARF: A Fast Deformer for Articulated Neural Fields [92.68788512596254]
  We propose a new articulation module for neural fields, Fast-SNARF, which finds accurate correspondences between canonical space and posed space. Fast-SNARF is a drop-in replacement in to our previous work, SNARF, while significantly improving its computational efficiency. Because learning of deformation maps is a crucial component in many 3D human avatar methods, we believe that this work represents a significant step towards the practical creation of 3D virtual humans.
  arXiv  Detail & Related papers  (2022-11-28T17:55:34Z)
• Rethinking IoU-based Optimization for Single-stage 3D Object Detection [103.83141677242871]
  We propose a Rotation-Decoupled IoU (RDIoU) method that can mitigate the rotation-sensitivity issue. Our RDIoU simplifies the complex interactions of regression parameters by decoupling the rotation variable as an independent term.
  arXiv  Detail & Related papers  (2022-07-19T15:35:23Z)
• The Devil is in the Pose: Ambiguity-free 3D Rotation-invariant Learning via Pose-aware Convolution [18.595285633151715]
  We develop a Pose-aware Rotation Invariant Convolution (i.e., PaRI-Conv) We propose an Augmented Point Pair Feature (APPF) to fully encode the RI relative pose information, and a factorized dynamic kernel for pose-aware kernel generation. Our PaRI-Conv surpasses the state-of-the-art RI methods while being more compact and efficient.
  arXiv  Detail & Related papers  (2022-05-30T16:11:55Z)
• Learning Local Neighboring Structure for Robust 3D Shape Representation [143.15904669246697]
  Representation learning for 3D meshes is important in many computer vision and graphics applications. We propose a local structure-aware anisotropic convolutional operation (LSA-Conv) Our model produces significant improvement in 3D shape reconstruction compared to state-of-the-art methods.
  arXiv  Detail & Related papers  (2020-04-21T13:40:03Z)

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.