OrthoMotion:Disentangling Camera and Subject Motion via Geometry Semantics Orthogonal Attention

• URL: http://arxiv.org/abs/2606.22835v1
• Date: Mon, 22 Jun 2026 04:15:14 GMT
• Title: OrthoMotion:Disentangling Camera and Subject Motion via Geometry Semantics Orthogonal Attention
• Abstract summary: Controllable video generation demands independent command of the camera and the subject.<n> OrthoMotion routes camera motion into a geometric channel, a norm-preserving rotation of the rotary position embedding (RoPE) phase, and subject motion into a semantic channel.<n>To our knowledge OrthoMotion is the first method to guarantee disentanglement by construction rather than hope for it to emerge.
• Score: 0.4790056963046066
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Controllable video generation demands independent command of the camera and the subject, yet 2D conditioning entangles them: camera- and object-induced optical flow share the same inverse-depth (1/Z) scaling and cannot be separated from image evidence alone. We first prove that this entanglement is representational, not architectural -- the 2D camera/object split is a non-identifiable inverse problem -- and therefore reframe decoupling as a question of operator design. We resolve it at the level of the attention operator. OrthoMotion routes camera motion into a geometric channel, a norm-preserving rotation of the rotary position embedding (RoPE) phase, and subject motion into a semantic channel, a gated value injection in cross-attention. Because these sub-operators are algebraically complementary -- a rotation versus a translation of the affine action on tokens -- a lightweight decoupling regularizer provably drives their response subspaces to orthogonality, so the two controls stop interfering. To our knowledge OrthoMotion is the first method to guarantee disentanglement by construction rather than hope for it to emerge. It attains state-of-the-art camera and subject accuracy at once while minimizing cross-talk, which we quantify with a new Cross-Talk Error (CTE) metric, cutting cross-talk by more than 2.4x with no loss in fidelity and generalizing across backbones.

Related papers

• OmniPoint: Universal Monocular Metric Pointcloud from Any Camera [79.24417597712802]
  We present OmniPoint, a unified framework designed to generalize metric reconstruction across diverse imaging sensors.<n>To overcome projection rigidity, OmniPoint abandons conventional planar depth regression.<n>It instead adopts a decoupled ray and distance representation alongside a decoupled training objective.
  arXiv  Detail & Related papers  (2026-09-08T19:45:58Z)
• MeRoPE: Metric Rotary Position Embedding for Camera-Controlled Video Generation [29.69634930299998]
  MeRoPE (Metric Rotary Position Embedding) is a norm-preserving relative camera encoding for attention.<n>Across nuScenes and PanShot, MeRoPE achieves stronger camera control than prior encodings.
  arXiv  Detail & Related papers  (2026-09-01T13:48:31Z)
• RASA: Disentangled Spatial-Motional Priors for Cross-Identity Character Animation [85.41886214825894]
  Cross-identity character animation aims to drive a target identity from a reference image to follow the motion of a source character from a driving video.<n>We introduce Reference-Aware Structural Alignment (RASA), a framework that disentangles spatial mapping from motion control.<n>Our work establishes a new paradigm showing disentangled spatial and motional priors are key to robust character animation.
  arXiv  Detail & Related papers  (2026-08-28T11:36:44Z)
• Pix2Act: Image-Space Manipulation Policies with Equivariant Augmentation [42.33906056027883]
  Representing manipulation actions as 2D trajectories in the camera plane provides a compact and interpretable basis for learning complex 3D manipulation policies.<n>We propose Pix2Act, an imitation learning method that addresses these challenges by generating continuous image-space keypoint trajectories in each camera plane.<n>This reformulates high-dimensional 3D control as a simpler, more learnable 2D prediction problem.
  arXiv  Detail & Related papers  (2026-07-13T07:03:29Z)
• MoCapAnything V2: End-to-End Motion Capture for Arbitrary Skeletons [56.68975315643491]
  We present the first fully end-to-end framework in which Videoto-Pose and Pose-to-Rotation are learnable and optimized jointly.<n>We show that our method reduces rotation error from 17 degrees to 10 degrees, and to 6.54 degrees on unseen skeletons, while achieving 20x faster inference than mesh-based pipelines.
  arXiv  Detail & Related papers  (2026-04-30T17:16:38Z)
• FLIGHT: Fibonacci Lattice-based Inference for Geometric Heading in real-Time [7.517221623631364]
  Estimating camera motion from monocular video is a fundamental problem in computer vision.<n>Existing methods that recover the camera's heading under known rotation tend to perform well in low-noise, low-outlier conditions.<n>We propose a novel generalization of the Hough transform on the unit sphere to estimate the camera's heading.
  arXiv  Detail & Related papers  (2026-02-26T15:27:49Z)
• DualCamCtrl: Dual-Branch Diffusion Model for Geometry-Aware Camera-Controlled Video Generation [51.66285725139235]
  We present DualCamCtrl, a novel end-to-end diffusion model for camera-controlled video generation.<n>We propose a dual-branch framework that mutually generates camera-consistent RGB and depth sequences.<n>Experiments demonstrate that DualCamCtrl achieves more consistent camera-controlled video generation.
  arXiv  Detail & Related papers  (2025-11-28T12:19:57Z)
• Look Before You Fuse: 2D-Guided Cross-Modal Alignment for Robust 3D Detection [7.448164560761331]
  Existing methods suffer from spatial misalignment between LiDAR and camera features.<n>The root cause of this misalignment lies in projection errors, stemming from calibration inaccuracies and rolling shutter effect.<n>We introduce Discontinuity Aware Geometric Fusion to suppress residual noise from PGDC and explicitly enhance sharp depth transitions at object-background boundaries.<n>Our method achieves SOTA performance on nuScenes validation dataset, with its mAP and NDS reaching 71.5% and 73.6% respectively.
  arXiv  Detail & Related papers  (2025-07-21T18:12:22Z)
• Single-Scanline Relative Pose Estimation for Rolling Shutter Cameras [56.39904484784127]
  We propose an approach for estimating the relative pose between rolling shutter cameras using the intersections of line projections with a single scanline per image.<n>Alternatively, scanlines can be selected within a single image, enabling single-view relative pose estimation for scanlines of rolling shutter cameras.
  arXiv  Detail & Related papers  (2025-06-27T10:00:21Z)
• 3D Trajectory Reconstruction of Moving Points Based on Asynchronous Cameras [6.9017898687323775]
  Localizing moving targets is crucial for analyzing their motion characteristics and dynamic properties.<n>This paper proposes a 3D trajectory reconstruction method for point targets based on asynchronous cameras.
  arXiv  Detail & Related papers  (2025-05-31T13:04:31Z)
• DFR: Depth from Rotation by Uncalibrated Image Rectification with Latitudinal Motion Assumption [6.369764116066747]
  We propose Depth-from-Rotation (DfR), a novel image rectification solution for uncalibrated rotating cameras. Specifically, we model the motion of a rotating camera as the camera rotates on a sphere with fixed latitude. We derive a 2-point analytical solver from directly computing the rectified transformations on the two images.
  arXiv  Detail & Related papers  (2023-07-11T09:11:22Z)
• Attentive and Contrastive Learning for Joint Depth and Motion Field Estimation [76.58256020932312]
  Estimating the motion of the camera together with the 3D structure of the scene from a monocular vision system is a complex task. We present a self-supervised learning framework for 3D object motion field estimation from monocular videos.
  arXiv  Detail & Related papers  (2021-10-13T16:45:01Z)
• Do Generative Models Know Disentanglement? Contrastive Learning is All You Need [59.033559925639075]
  We propose an unsupervised and model-agnostic method: Disentanglement via Contrast (DisCo) in the Variation Space. DisCo achieves the state-of-the-art disentanglement given pretrained non-disentangled generative models, including GAN, VAE, and Flow.
  arXiv  Detail & Related papers  (2021-02-21T08:01:20Z)

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.