Deformation-Free Cross-Domain Image Registration via Position-Encoded Temporal Attention

Yiwen Wang; Jiahao Qin

cs.CVFeb 17, 2026

Deformation-Free Cross-Domain Image Registration via Position-Encoded Temporal Attention

Authors: Yiwen Wang, Jiahao Qin

Abstract

We address the problem of cross-domain image registration, where paired images exhibit coupled geometric misalignment and domain-specific appearance shift. We formalize this as a factorization problem: decomposing each image into a domain-invariant scene representation and a global appearance statistic, such that registration reduces to recombining the scene structure of the moving image with the appearance of the fixed image via Adaptive Instance Normalization (AdaIN). This factorization eliminates the need for explicit deformation field estimation. To exploit temporal coherence in sequential acquisitions, we introduce a position-encoded cross-frame attention mechanism that fuses learnable and sinusoidal position embeddings with multi-head attention over a sliding window of neighboring frames, enriching the scene representation with inter-frame context. We instantiate this framework as GPEReg-Net and evaluate on two benchmarks: FIRE-Reg-256 (retinal fundus, semi-rigid) and HPatches-Reg-256 (synthetic textured patches, affine). GPEReg-Net achieves state-of-the-art performance on both benchmarks (FIRE: SSIM = 0.928, PSNR = 33.47 dB; HPatches: SSIM = 0.450, PSNR = 21.01 dB), surpassing all baselines, including deformation-based methods, while running 1.87x faster than SAS-Net. Code: https://github.com/JiahaoQin/GPEReg-Net.

Figures & tables

Panel of Figure 1 — Figure 1: (a) Registration example on HPatches-Reg-256 [ 3 ] : the fixed image $I_{f}$ , moving image $I_{m}$ with geometric misalignment, the registered output $\hat{I}_{r}$ produced by GPEReg-Net, and the absolute difference $|\hat{I}_{r}-I_{f}|$ . A dashed arrow indicates the appearance pathway from $I_{f}$ . (b) Overview of GPEReg-Net. The SceneEncoder $\mathcal{S}$ extracts domain-invariant features $\bm{s}$ from the moving image $I_{m}$ using instance normalization. The AppearanceEncoder $\mathcal{A}$ extracts a 32-dim global appearance code $\bm{a}$ from the fixed image $I_{f}$ . The GPE module $\mathcal{G}$ enriches scene features with temporal context via learnable position embeddings, sinusoidal encoding, and cross-frame attention, producing $\tilde{\bm{s}}$ . The ImageDecoder $\mathcal{D}$ uses AdaIN to modulate enhanced scene features with the target appearance, producing the registered output $\hat{I}_{r}$ without any explicit deformation field.

Traditional Methods
Method	NCC $\uparrow$	SSIM $\uparrow$	PSNR $\uparrow$
Unregistered	0.762	0.494	22.36
SIFT [ 12 ]	0.449	0.463	16.39
Demons [ 27 ]	0.672	0.528	17.45
Optical Flow [ 9 ]	0.552	0.506	16.77
SyN [ 1 ]	0.549	0.521	15.76

Table 1: Registration on FIRE-Reg-256 [ 8 ] (973 test patches). Our method in bold .

Method	Params (M)	Latency (ms)	FPS
VoxelMorph [ 2 ]	0.10	3.06	327
TransMorph [ 4 ]	0.17	2.85	351
SAS-Net [ 23 ]	3.35	27.21	37
GPEReg-Net (Ours)	3.40	14.52	69

Table 2: Computational efficiency comparison. Parameters (M), inference latency (ms), and throughput (FPS) benchmarked on an RTX 5090 GPU.

Method	NCC $\uparrow$	SSIM $\uparrow$	PSNR $\uparrow$
Unregistered	0.312	0.241	14.87
VoxelMorph [ 2 ]	0.448	0.376	18.53
TransMorph [ 4 ]	0.471	0.395	19.24
SAS-Net [ 23 ]	0.502	0.421	20.15
GPEReg-Net (Ours)	0.536	0.450	21.01

Table 3: Cross-domain applicability on HPatches-Reg-256 [ 3 ] (500 test pairs). All deep learning methods are trained from scratch on HPatches-Reg-256 for 30 epochs. Best in bold .

Figure 2: Qualitative registration results on HPatches-Reg-256. Four representative test samples showing fixed, moving, registered output, and absolute difference map with per-sample SSIM annotations.

Figure 3: Red-green overlay comparison on HPatches-Reg-256 before and after registration (red = fixed, green = moving/registered). Reduced color separation indicates better spatial alignment.

Temporal Attention

Temporal attention models how information across time should be selected and combined in video, sequences, and time-series data.

See all 64 papers

Temporal Periodic Image Registration with Implicit Neural Representations
Oct 15, 2026Mathias Lowes, Kristine Aavild Sørensen, Maxime Sermesant +1Deformable Image Registration Medical Image Registration
Department of Applied Mathematics and Computer Science, Technical University of Denmark, Kgs. Lyngby, Denmark · Novo Nordisk A/S, Copenhagen, Denmark · Inria Université Côte d’Azur, Nice, France
CRFT: Consistent-Recurrent Feature Flow Transformer for Cross-Modal Image Registration
Apr 7, 2026Xuecong Liu, Mengzhu Ding, Zixuan Sun +2Cross-Modal Image Registration Cross-Modal Feature Matching
Northeastern University, China · National University of Defense Technology, China
OCA: ODE-Driven Cross-Attention for Image-to-Point-Cloud Registration
Sep 29, 2026Pei An, Jiaqi Yang, Yulong Wang +2Cross-Attention Point Cloud Registration
Huazhong University of Science and Technology, China · Northwestern Polytechnical University, China · Huazhong Agricultural University, China +2

Oct 15, 2026J/K move · Enter open · S save

Temporal Periodic Image Registration with Implicit Neural Representations

Mathias Lowes, Kristine Aavild Sørensen, Maxime Sermesant, Rasmus R. Paulsen

Department of Applied Mathematics and Computer Science, Technical University of Denmark, Kgs. Lyngby, Denmark · Novo Nordisk A/S, Copenhagen, Denmark · Inria Université Côte d’Azur, Nice, France

Deformable Image Registration Medical Image Registration Implicit Neural Representations

Open paper AskPDF arXiv

Implicit Neural Representations (INRs) have recently gained popularity for their ability to model functions using simple and lightweight networks. These methods have also proven effective in deformable image registration. In this work, we extend INRs for deformable image registration to the domain of temporal image registration, focusing on periodic temporal image sequences. Our approach, Temporal-IDIR, optimizes a single INR to model deformations across all frames in a temporal image sequence simultaneously, allowing for self-regularization through its own deformation predictions. To achieve this, we introduce a temporal consistency loss that penalizes discrepancies between direct source-to-target transformations and those traversing intermediate frames. We evaluate our framework on the DIR-LAB dataset, using the target registration error (TRE) between annotated and moved landmarks as the metric. Here, we achieve a TRE of 1.03 mm, outperforming other INR-based registration methods. Additionally, our framework supports smooth interpolation between time frames by estimating deformations between the given input frames. Code is publicly available at https://github.com/MMLowes/Temporal_INR.

Abstract

Temporal Attention

Explore similar work

Temporal Periodic Image Registration with Implicit Neural Representations

CRFT: Consistent-Recurrent Feature Flow Transformer for Cross-Modal Image Registration

OCA: ODE-Driven Cross-Attention for Image-to-Point-Cloud Registration