Who Handles Orientation? Investigating Invariance in Feature Matching

• URL: http://arxiv.org/abs/2604.11809v1
• Date: Mon, 13 Apr 2026 17:59:58 GMT
• Title: Who Handles Orientation? Investigating Invariance in Feature Matching
• Abstract summary: We release two matchers robust to in-plane rotations that achieve state-of-the-art performance on e.g. multi-modal (WxBS), extreme (HardMatch), and satellite image matching (SatAst)
• Score: 29.044546222577804
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Finding matching keypoints between images is a core problem in 3D computer vision. However, modern matchers struggle with large in-plane rotations. A straightforward mitigation is to learn rotation invariance via data augmentation. However, it remains unclear at which stage rotation invariance should be incorporated. In this paper, we study this in the context of a modern sparse matching pipeline. We perform extensive experiments by training on a large collection of 3D vision datasets and evaluating on popular image matching benchmarks. Surprisingly, we find that incorporating rotation invariance already in the descriptor yields similar performance to handling it in the matcher. However, rotation invariance is achieved earlier in the matcher when it is learned in the descriptor, allowing for a faster rotation-invariant matcher. Further, we find that enforcing rotation invariance does not hurt upright performance when trained at scale. Finally, we study the emergence of rotation invariance through scale and find that increasing the training data size substantially improves generalization to rotated images. We release two matchers robust to in-plane rotations that achieve state-of-the-art performance on e.g. multi-modal (WxBS), extreme (HardMatch), and satellite image matching (SatAst). Code is available at https://github.com/davnords/loma.

Related papers

• Steerers: A framework for rotation equivariant keypoint descriptors [26.31402935889126]
  Keypoint descriptions that are discriminative and matchable over large changes in viewpoint are vital for 3D reconstruction. We learn a linear transform in description space that encodes rotations of the input image. We obtain state-of-the-art results on the rotation invariant image matching benchmarks AIMS and Roto-360.
  arXiv  Detail & Related papers  (2023-12-04T18:59:44Z)
• Evaluating 3D Shape Analysis Methods for Robustness to Rotation Invariance [22.306775502181818]
  This paper analyzes the robustness of recent 3D shape descriptors to SO(3) rotations. We consider a database of 3D indoor scenes, where objects occur in different orientations.
  arXiv  Detail & Related papers  (2023-05-29T18:39:31Z)
• Sorted Convolutional Network for Achieving Continuous Rotational Invariance [56.42518353373004]
  We propose a Sorting Convolution (SC) inspired by some hand-crafted features of texture images. SC achieves continuous rotational invariance without requiring additional learnable parameters or data augmentation. Our results demonstrate that SC achieves the best performance in the aforementioned tasks.
  arXiv  Detail & Related papers  (2023-05-23T18:37:07Z)
• Rotation-Invariant Transformer for Point Cloud Matching [42.5714375149213]
  We introduce RoITr, a Rotation-Invariant Transformer to cope with the pose variations in the point cloud matching task. We propose a global transformer with rotation-invariant cross-frame spatial awareness learned by the self-attention mechanism. RoITr surpasses the existing methods by at least 13 and 5 percentage points in terms of Inlier Ratio and Registration Recall.
  arXiv  Detail & Related papers  (2023-03-14T20:55:27Z)
• General Rotation Invariance Learning for Point Clouds via Weight-Feature Alignment [40.421478916432676]
  We propose Weight-Feature Alignment (WFA) to construct a local Invariant Reference Frame (IRF) Our WFA algorithm provides a general solution for the point clouds of all scenes.
  arXiv  Detail & Related papers  (2023-02-20T11:08:07Z)
• RIC-CNN: Rotation-Invariant Coordinate Convolutional Neural Network [56.42518353373004]
  We propose a new convolutional operation, called Rotation-Invariant Coordinate Convolution (RIC-C) By replacing all standard convolutional layers in a CNN with the corresponding RIC-C, a RIC-CNN can be derived. It can be observed that RIC-CNN achieves the state-of-the-art classification on the rotated test dataset of MNIST.
  arXiv  Detail & Related papers  (2022-11-21T19:27:02Z)
• ART-Point: Improving Rotation Robustness of Point Cloud Classifiers via Adversarial Rotation [89.47574181669903]
  In this study, we show that the rotation robustness of point cloud classifiers can also be acquired via adversarial training. Specifically, our proposed framework named ART-Point regards the rotation of the point cloud as an attack. We propose a fast one-step optimization to efficiently reach the final robust model.
  arXiv  Detail & Related papers  (2022-03-08T07:20:16Z)
• Extreme Rotation Estimation using Dense Correlation Volumes [73.35119461422153]
  We present a technique for estimating the relative 3D rotation of an RGB image pair in an extreme setting. We observe that, even when images do not overlap, there may be rich hidden cues as to their geometric relationship. We propose a network design that can automatically learn such implicit cues by comparing all pairs of points between the two input images.
  arXiv  Detail & Related papers  (2021-04-28T02:00:04Z)
• Adjoint Rigid Transform Network: Task-conditioned Alignment of 3D Shapes [86.2129580231191]
  Adjoint Rigid Transform (ART) Network is a neural module which can be integrated with a variety of 3D networks. ART learns to rotate input shapes to a learned canonical orientation, which is crucial for a lot of tasks. We will release our code and pre-trained models for further research.
  arXiv  Detail & Related papers  (2021-02-01T20:58:45Z)
• What Does CNN Shift Invariance Look Like? A Visualization Study [87.79405274610681]
  Feature extraction with convolutional neural networks (CNNs) is a popular method to represent images for machine learning tasks. We focus on measuring and visualizing the shift invariance of extracted features from popular off-the-shelf CNN models. We conclude that features extracted from popular networks are not globally invariant, and that biases and artifacts exist within this variance.
  arXiv  Detail & Related papers  (2020-11-09T01:16:30Z)
• Rotated Ring, Radial and Depth Wise Separable Radial Convolutions [13.481518628796692]
  In this work, we address trainable rotation invariant convolutions and the construction of nets. On the one hand, we show that our approach is rotationally invariant for different models and on different public data sets. The rotationally adaptive convolution models presented are more computationally intensive than normal convolution models.
  arXiv  Detail & Related papers  (2020-10-02T09:01:51Z)

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.