SemGeo-Gen: Unsupervised Generation of Approximate Cross-Instance Semantic-Geometric Correspondences

Roy Amoyal; Shira Ifergane; Oren Freifeld

cs.CVOct 23, 2026

SemGeo-Gen: Unsupervised Generation of Approximate Cross-Instance Semantic-Geometric Correspondences

Authors: Roy Amoyal, Shira Ifergane, Oren Freifeld

Organizations: The Stein Faculty of Computer and Information Science, Ben-Gurion University of the Negev, Israel

Abstract

We introduce SemGeo-Gen, an unsupervised method for generating approximate cross-instance semantic-geometric supervision from weakly aligned 3D object collections. Our goal is not to recover exact point-level correspondences, which are often ambiguous across different instances and would require expensive, impractical manual supervision at scale. Instead, we automatically produce approximate correspondences that are semantically meaningful, geometrically consistent, and diverse enough to train modern correspondence models. Given only coarse category-level rotation alignment, SemGeo-Gen lifts multi-view DINOv2 features to 3D, aligns object instances using continuous piecewise-affine registration, and prunes candidate matches using semantic consistency. The resulting 3D correspondences can be projected into rendered views, yielding scalable 3D-3D, 3D-2D, and 2D-2D supervision without manual annotation. We validate the generated correspondences against sparse human annotations from KeypointNet, obtaining 93% [email protected] and 97% [email protected]. More importantly, we show that using the generated correspondences for synthetic pretraining improves recent state-of-the-art semantic correspondence models on SPair-71k after fine-tuning on real data, with gains of up to 5 [email protected] points. Together, these results indicate that approximate semantic-geometric supervision generated from 3D assets can improve real-image correspondence learning and serve as a scalable alternative to costly manual annotation. Project page: https://bgu-cs-vil.github.io/SemGeo-Gen/.

Figures & tables

Panel of Figure 1 — Figure 1: Approximate cross-instance semantic-geometric correspondences between two airplane instances across modalities. Left: 2D–2D correspondences between rendered views. Middle: 3D–2D correspondences between a 3D object and a rendered view. Right: 3D–3D correspondences between object surfaces. Colors in 3D visualize the lifted DINOv2 features reduced to three dimensions.

Figure 2: Overview of our correspondence generation pipeline.

Zero-shot / weakly supervised methods
Method	Training regime	aero	bike	boat	bus	car	chair	mbike	train	tv	Mean
SD [ 27 , 38 ]	Zero-shot	62.8	52.7	31.2	39.1	35.6	32.0	51.0	61.8	52.9	46.57
DINOv2 [ 25 , 38 ]	Zero-shot	73.4	60.2	43.2	46.7	45.1	33.4	60.7	54.2	23.9	48.98
DINOv2+SD [ 38 ]	Zero-shot	73.8	61.0	40.2	47.4	44.1	41.5	61.7	63.5	52.4	53.96
SphericalMaps [ 21 ]	Weakly supervised (SPair-71k)	76.2	60.1	46.5	74.9	68.0	45.1	69.1	73.9	58.1	63.54
Supervised methods

Table 1: Category-wise [email protected] results on selected SPair-71k categories with well-defined semantic-geometric correspondences. P.T. denotes pretraining.

Training regime	Rigid-9	Non-rigid-7	Sym.-2	All-18
Real only	80.48	86.94	72.35	82.09
DenseMatcher [ 40 ] P.T. $\rightarrow$ Real	80.37	86.54	71.97	81.83
Ours P.T. $\rightarrow$ Real	84.20	88.23	72.58	84.48
	(+3.72)	(+1.29)	(+0.23)	(+2.39)

Table 2: Generalization across all SPair-71k categories. GeoAware-SC pretrained on Rigid-9 synthetic correspondences and fine-tuned on all 18 SPair-71k categories. [email protected] (%); improvements over real-only training are shown in parentheses.

Appendix figures & tables33 assets

Supplementary material from the paper’s appendix.

Appendix

Panel of Figure 6 — Figure 6: Qualitative SPair-71k [ 23 ] results for GeoAware-SC [ 39 ] . Each image shows a source–target pair with predicted correspondences. We compare GeoAware-SC trained on SPair-71k against GeoAware-SC pretrained on our generated correspondences and then fine-tuned on SPair-71k.

Panel of Figure 7 — Figure 7: Qualitative correspondence examples on categories beyond existing correspondence benchmarks. Colors indicate sampled semantic-geometric correspondences established by our method.

Panel of Figure 8 — Figure 8: Qualitative correspondence results on KeypointNet [ 35 ] We compare our predicted correspondences with human-annotated keypoints on the target object. Predicted correspondences are shown in lighter colors, while human annotations are shown in darker colors. Red lines connect each prediction to its corresponding ground-truth annotation, visualizing the localization difference between them.

Panel of Figure 9 — Figure 9: KeypointNet [ 35 ] examples with strong annotation agreement. Predicted correspondences are shown in lighter colors, while human annotations are shown in darker colors. The strong overlap between the two indicates that our pipeline recovers correspondences that are highly consistent with human semantic annotations.

Method	@0.05	@0.10	@0.15	@0.20
Back to 3D (1-shot)	0.438	0.632	0.725	0.777
Back to 3D (3-shot)	0.436	0.641	0.740	0.799
SemGeo-Gen	0.648	0.865	0.945	0.977

Appendix

Table 3: Comparison with Back to 3D [ 33 ] on KeypointNet [ 35 ] Mean PCK over 16 categories (100 pairs per category). Back to 3D uses few-shot keypoint supervision; SemGeo-Gen requires none.

Method	DOF	Time (s/pair)	[email protected]	[email protected]	[email protected]	[email protected]	[email protected]
Affine	12	1.15	0.221	0.686	0.898	0.960	0.982
TPS ( $K\!=\!216$ )	660	3.71	0.226	0.739	0.919	0.966	0.983
CPA (ours)	648	1.13	0.236	0.729	0.922	0.971	0.986

Appendix

Table 4: Registration ablation on KeypointNet. Mean PCK across all 16 categories. PCK thresholds define the allowed normalized distance from the ground-truth keypoint, measured as a fraction of the target bounding-box diagonal. DOF denotes the number of learnable parameters. Time reports the mean registration time per pair on a single A6000 GPU. Best results are shown in bold .

Category	[email protected] $\uparrow$	[email protected] $\uparrow$	[email protected] $\uparrow$	[email protected] $\uparrow$
Airplane	0.73	0.90	0.95	0.98
Bathtub	0.41	0.73	0.91	0.97
Bed	0.45	0.71	0.84	0.90
Bottle	0.65	0.83	0.92	0.98
Car	0.75	0.94	0.97	0.99
Chair	0.51	0.83	0.93	0.97

Appendix

Table 5: Evaluation of annotation quality on KeypointNet correspondences without filters. We report PCK at normalized distance thresholds of

5\%

10\%

15\%

, and

20\%

of the object scale, computed per category. (on 100 objects)

Category	[email protected]	[email protected]	[email protected]	[email protected]
Airplane	0.78	0.94	0.98	0.99
Bathtub	0.68	0.86	0.94	0.97
Bed	0.63	0.85	0.94	0.98
Bottle	0.80	0.95	0.97	0.99
Car	0.77	0.95	0.98	1.00
Chair	0.71	0.94	0.98	0.99

Appendix

Table 6: Ablation of only semantic filtering of correspondences, reporting PCK results per category. calculated over 100 models.

$\tau_{\mathrm{sem}}$	[email protected] $\uparrow$	[email protected] $\uparrow$	[email protected] $\uparrow$	[email protected] $\uparrow$	Median GT-used rate $\uparrow$
0.75	0.758	0.930	0.974	0.988	0.260
0.575	0.749	0.926	0.973	0.989	0.576
0.25	0.699	0.896	0.960	0.983	0.915

Appendix

Table 7: Ablation on the correspondence-level semantic threshold (with pair-level filtering). Results are averaged over three random seeds and 16 KeypointNet categories. Median GT-used rate denotes the median fraction across categories of ground-truth keypoints that remain evaluable after correspondence-level pruning.

PCK	Stat.	$0^{\circ}$	$5^{\circ}$	$10^{\circ}$	$30^{\circ}$	$60^{\circ}$	$90^{\circ}$
@0.05	Mean	.75	.71	.65	.32	.10	.04
@0.05	Median	.78	.75	.69	.25	.06	.02
@0.10	Mean	.93	.89	.85	.55	.20	.08
@0.10	Median	.94	.91	.88	.52	.13	.05
@0.15	Mean	.98	.94	.90	.66	.26	.12
@0.15	Median	.98	.95	.91	.64	.20	.07

Appendix

Table 8: Sensitivity to rotation misalignment. Mean and median KeypointNet PCK under random-axis rotations applied to either the source or target.

Panel of Figure 10 — Figure 10: Qualitative analysis of DINOv2 based semantic coverage. Each row shows the original rendering, a PCA visualization of DINOv2 features computed jointly for the object pair, and the resulting semantic coverage mask. Green points have at least one semantically similar point in the paired object, while red points have none.

Panel of Figure 11 — Figure 11: Sampled correspondences after semantic pruning. Each row shows a different view of the same object pair.

Category	[email protected]	[email protected]	[email protected]	[email protected]	Coverage	3D [email protected]
Airplane	0.79	0.94	0.98	0.99	0.73	0.94
Bathtub	0.81	0.95	0.99	1.00	0.26	0.88
Bed	0.81	0.96	0.99	1.00	0.45	0.91
Bottle	0.81	0.95	0.98	0.99	0.40	0.96
Cap	0.83	0.97	1.00	1.00	0.30	0.97
Car	0.76	0.95	0.98	1.00	0.88	0.95

Appendix

Table 9: 3D–2D correspondence evaluation on KeypointNet. PCK on found correspondences with visible ground-truth keypoints and visible predicted target points. Coverage denotes the fraction of visible ground-truth keypoints for which a valid correspondence is found. The last column reports the original 3D [email protected] over all found keypoints.

ID	Description	[email protected]	[email protected]	[email protected]
0	nose of the airplane	$3805/3834$	$3829/3834$	$3834/3834$
1	cockpit of the airplane	$2768/3289$	$3113/3289$	$3205/3289$
2	tail of the airplane	$3059/3457$	$3437/3457$	$3455/3457$
3	wingtip of the left wing of the airplane	$3597/3790$	$3707/3790$	$3716/3790$
4	wing root of the left wing of the airplane	$3046/3192$	$3181/3192$	$3192/3192$
5	wingtip of the right wing of the airplane	$3606/3795$	$3709/3795$	$3719/3795$

Appendix

Table 10: Per-keypoint PCK results for category airplane . Values shown as

\text{correct}/\text{total}

ID	Description	[email protected]	[email protected]	[email protected]
0	top outer corner of the bathtub	$1340/1350$	$1346/1350$	$1350/1350$
1	top outer corner of the bathtub	$1300/1314$	$1310/1314$	$1314/1314$
2	top outer corner of the bathtub	$1277/1287$	$1285/1287$	$1287/1287$
3	top outer corner of the bathtub	$1234/1250$	$1248/1250$	$1250/1250$
4	bottom outer corner of the bathtub	$566/649$	$595/649$	$616/649$
5	bottom outer corner of the bathtub	$478/558$	$511/558$	$530/558$

Appendix

Table 11: Per-keypoint PCK results for category bathtub . Values shown as

\text{correct}/\text{total}

ID	Description	[email protected]	[email protected]	[email protected]
0	lower end of the front leg of the bed	$989/1044$	$1032/1044$	$1042/1044$
1	lower end of the headboard leg of the bed	$880/965$	$924/965$	$960/965$
2	lower end of the headboard leg of the bed	$825/906$	$857/906$	$901/906$
3	lower end of the front leg of the bed	$937/988$	$979/988$	$984/988$
4	foot corner of the mattress	$606/663$	$640/663$	$648/663$
5	head corner of the mattress	$420/511$	$464/511$	$491/511$

Appendix

Table 12: Per-keypoint PCK results for category bed . Values shown as

\text{correct}/\text{total}

ID	Description	[email protected]	[email protected]	[email protected]
0	lower rim of the bottle	$2836/2866$	$2860/2866$	$2864/2866$
1	lower rim of the bottle	$2845/2868$	$2866/2868$	$2866/2868$
2	lower rim of the bottle	$2693/2711$	$2707/2711$	$2711/2711$
3	lower rim of the bottle	$2829/2861$	$2857/2861$	$2860/2861$
4	edge between the bottle body and the bottle neck	$243/261$	$259/261$	$261/261$
5	edge between the bottle body and the bottle neck	$257/275$	$272/275$	$275/275$

Appendix

Table 13: Per-keypoint PCK results for category bottle . Values shown as

\text{correct}/\text{total}

ID	Description	[email protected]	[email protected]	[email protected]
0	top button of the cap	$528/558$	$557/558$	$558/558$
1	front of the visor of the cap	$558/558$	$558/558$	$558/558$
2	corner of the visor of the cap	$481/558$	$546/558$	$554/558$
3	corner of the visor of the cap	$484/558$	$540/558$	$555/558$
4	center of the back of the cap	$555/558$	$558/558$	$558/558$
5	center of the visor attachment of the cap	$493/502$	$502/502$	$502/502$

Appendix

Table 14: Per-keypoint PCK results for category cap . Values shown as

\text{correct}/\text{total}

ID	Description	[email protected]	[email protected]	[email protected]
0	front right wheel of the car	$3118/3118$	$3118/3118$	$3118/3118$
1	rear right wheel of the car	$3097/3100$	$3100/3100$	$3100/3100$
2	rear left wheel of the car	$3102/3113$	$3113/3113$	$3113/3113$
3	front left wheel of the car	$3107/3111$	$3111/3111$	$3111/3111$
4	corner of the front bumper of the car	$2890/2909$	$2909/2909$	$2909/2909$
5	corner of the rear bumper of the car	$3046/3067$	$3067/3067$	$3067/3067$

Appendix

Table 15: Per-keypoint PCK results for category car . Values shown as

\text{correct}/\text{total}

ID	Description	[email protected]	[email protected]	[email protected]
0	upper corner of the back of the chair	$743/779$	$771/779$	$778/779$
1	upper corner of the back of the chair	$713/759$	$757/759$	$758/759$
2	lower corner of the back of the chair	$501/548$	$537/548$	$547/548$
3	lower corner of the back of the chair	$489/539$	$526/539$	$537/539$
4	front corner of the seat of the chair	$822/840$	$839/840$	$840/840$
5	front corner of the seat of the chair	$816/840$	$839/840$	$840/840$

Appendix

Table 16: Per-keypoint PCK results for category chair . Values shown as

\text{correct}/\text{total}

ID	Description	[email protected]	[email protected]	[email protected]
0	head of the guitar	$3885/3904$	$3904/3904$	$3904/3904$
1	connection of the body and neck of the guitar	$2690/2926$	$2883/2926$	$2916/2926$
2	corner of the body of the guitar	$1964/2039$	$2014/2039$	$2033/2039$
3	corner of the body of the guitar	$2046/2060$	$2057/2060$	$2060/2060$
4	corner of the body of the guitar	$2104/2135$	$2128/2135$	$2135/2135$
5	corner of the body of the guitar	$2038/2130$	$2112/2130$	$2130/2130$

Appendix

Table 17: Per-keypoint PCK results for category guitar . Values shown as

\text{correct}/\text{total}

ID	Description	[email protected]	[email protected]	[email protected]
0	top of the helmet	$1765/1976$	$1949/1976$	$1974/1976$
1	front of the helmet peak	$1331/1976$	$1694/1976$	$1872/1976$
2	back of the helmet peak	$1451/1976$	$1789/1976$	$1915/1976$
3	left side of the helmet peak	$1300/1976$	$1738/1976$	$1895/1976$
4	right side of the helmet peak	$1343/1976$	$1762/1976$	$1912/1976$
5	right bottom of the helmet	$1296/1834$	$1592/1834$	$1697/1834$

Appendix

Table 18: Per-keypoint PCK results for category helmet . Values shown as

\text{correct}/\text{total}

ID	Description	[email protected]	[email protected]	[email protected]
0	corner of the laptop body	$2862/2870$	$2870/2870$	$2870/2870$
1	hinge of the laptop	$2334/2342$	$2342/2342$	$2342/2342$
2	hint of the laptop	$2193/2223$	$2222/2223$	$2223/2223$
3	corner of the laptop body	$2774/2780$	$2780/2780$	$2780/2780$
4	corner of the laptop screen	$2778/2791$	$2791/2791$	$2791/2791$
5	corner of the laptop screen	$2763/2770$	$2770/2770$	$2770/2770$

Appendix

Table 19: Per-keypoint PCK results for category laptop . Values shown as

\text{correct}/\text{total}

ID	Description	[email protected]	[email protected]	[email protected]
0	headlight of the motorcycle	$2392/2875$	$2788/2875$	$2864/2875$
1	end of the handlebar of the motorcycle	$3005/3549$	$3320/3549$	$3478/3549$
2	end of the handlebar of the motorcycle	$3009/3545$	$3301/3545$	$3481/3545$
3	seat of the motorcycle	$3763/4092$	$4020/4092$	$4087/4092$
4	axle of the front wheel of the motorcycle	$4225/4267$	$4264/4267$	$4267/4267$
5	axle of the back wheel of the motorcycle	$3873/3914$	$3911/3914$	$3914/3914$

Appendix

Table 20: Per-keypoint PCK results for category motorcycle . Values shown as

\text{correct}/\text{total}

ID	Description	[email protected]	[email protected]	[email protected]
0	mug rim	$1985/1996$	$1996/1996$	$1996/1996$
1	mug rim	$2038/2045$	$2044/2045$	$2045/2045$
2	mug rim	$2283/2307$	$2305/2307$	$2306/2307$
3	mug rim	$2274/2294$	$2293/2294$	$2294/2294$
4	mug base rim	$2718/2870$	$2843/2870$	$2868/2870$
5	mug base rim	$2733/2874$	$2845/2874$	$2874/2874$

Appendix

Table 21: Per-keypoint PCK results for category mug . Values shown as

\text{correct}/\text{total}

ID	Description	[email protected]	[email protected]	[email protected]
0	corner of the tail of the skateboard	$1308/1409$	$1380/1409$	$1404/1409$
1	corner of the tail of the skateboard	$1314/1421$	$1386/1421$	$1415/1421$
2	corner of the nose of the skateboard	$1289/1445$	$1404/1445$	$1440/1445$
3	corner of the nose of the skateboard	$1233/1395$	$1337/1395$	$1392/1395$
4	wheel of the skateboard	$2680/2694$	$2692/2694$	$2693/2694$
5	wheel of the skateboard	$2606/2622$	$2615/2622$	$2619/2622$

Appendix

Table 22: Per-keypoint PCK results for category skateboard . Values shown as

\text{correct}/\text{total}

ID	Description	[email protected]	[email protected]	[email protected]
0	corner of the table top	$1466/1523$	$1515/1523$	$1521/1523$
1	corner of the table top	$1604/1654$	$1646/1654$	$1652/1654$
2	corner of the table top	$1640/1676$	$1675/1676$	$1675/1676$
3	corner of the table top	$1638/1681$	$1670/1681$	$1674/1681$
4	end of the table leg	$2419/2584$	$2547/2584$	$2574/2584$
5	end of the table leg	$2443/2583$	$2552/2583$	$2580/2583$

Appendix

Table 23: Per-keypoint PCK results for category table . Values shown as

\text{correct}/\text{total}

ID	Description	[email protected]	[email protected]	[email protected]
0	bow of a boat	$1133/1142$	$1142/1142$	$1142/1142$
1	lower part of the end of the bow of a boat (starboard)	$396/668$	$553/668$	$634/668$
2	upper part of the end of the bow of a boat (starboard)	$179/292$	$245/292$	$276/292$
3	upper part of the end of the bow of a boat (port)	$213/323$	$277/323$	$313/323$
4	lower part of the end of the bow of a boat (port)	$445/717$	$611/717$	$684/717$
5	lower corner of the boat stern (starboard	$207/265$	$248/265$	$257/265$

Appendix

Table 24: Per-keypoint PCK results for category vessel . Values shown as

\text{correct}/\text{total}

Category	[email protected]	[email protected]	[email protected]	[email protected]
airplane	0.80	0.94	0.98	0.99
bathtub	0.72	0.89	0.96	0.99
bed	0.69	0.91	0.97	0.99
bottle	0.82	0.96	0.99	0.99
cap	0.80	0.97	1.00	1.00
car	0.78	0.96	0.99	1.00

Appendix

Table 25: PCK averaged over seeds, semantic_threshold = 0.625

Category	[email protected]	[email protected]	[email protected]	[email protected]
airplane	0.79	0.94	0.98	0.99
bathtub	0.70	0.88	0.96	0.99
bed	0.68	0.91	0.96	0.99
bottle	0.83	0.96	0.98	0.99
cap	0.80	0.97	1.00	1.00
car	0.78	0.95	0.99	1.00

Appendix

Table 26: PCK averaged over seeds, semantic_threshold = 0.575

Category	[email protected]	[email protected]	[email protected]	[email protected]
airplane	0.78	0.94	0.98	0.99
bathtub	0.68	0.87	0.96	0.99
bed	0.67	0.90	0.96	0.98
bottle	0.82	0.95	0.98	0.99
cap	0.79	0.97	1.00	1.00
car	0.78	0.95	0.98	0.99

Appendix

Table 27: PCK averaged over seeds, semantic_threshold = 0.525

Category	[email protected]	[email protected]	[email protected]	[email protected]
airplane	0.76	0.92	0.97	0.99
bathtub	0.53	0.79	0.93	0.98
bed	0.62	0.86	0.93	0.97
bottle	0.75	0.91	0.97	0.99
cap	0.79	0.95	0.99	1.00
car	0.77	0.95	0.98	0.99

Appendix

Table 28: PCK averaged over seeds, semantic_threshold = 0.25

Semantic Correspondence

Semantic correspondence matches equivalent points or parts across images and shapes despite changes in appearance, pose, or viewpoint.

See all 24 papers

Geometry Matters: 3D Foundation Priors for Learning Semantic Correspondence
May 28, 2026Artur Jesslen, Olaf Dünkel, Adam KortylewskiSemantic Correspondence
University of Freiburg, Germany · Max Planck Institute for Informatics, Saarland Informatics Campus, Germany · CISPA Helmholtz Center for Information Security, Germany
Shape-of-You: Fused Gromov-Wasserstein Optimal Transport for Semantic Correspondence in-the-Wild
Mar 12, 2026Jiin Im, Sisung Liu, Je Hyeong HongSemantic Correspondence Optimal Transport
Dept. Electronic Engineering, Hanyang University · Dept. Artificial Intelligence, Hanyang University
GeoWorld: Providing Full-frame Geometry Features to Facilitate 3D Scene Generation
Nov 28, 2025Yuhao Wan, Lijuan Liu, Jingzhi Zhou +6Video Diffusion Models 3D Scene Generation
1VCIP & AAIS, Nankai University · 2ByteDance Inc. · 3Renmin University of China +2

May 28, 2026J/K move · Enter open · S save

Geometry Matters: 3D Foundation Priors for Learning Semantic Correspondence

Artur Jesslen, Olaf Dünkel, Adam Kortylewski

University of Freiburg, Germany · Max Planck Institute for Informatics, Saarland Informatics Campus, Germany · CISPA Helmholtz Center for Information Security, Germany

Semantic Correspondence

Open paper AskPDF arXiv

Foundation features from self-supervised vision models and text-to-image diffusion models have proven effective for semantic correspondence estimation. However, because these features are learned primarily from 2D image objectives, they lack explicit 3D awareness and often confuse symmetric object sides, repeated parts, and visually similar structures that are distinct in 3D. We introduce a 3D-aware post-training framework that goes beyond available 2D foundation features by incorporating priors from 3D foundation models. Given an image, our method uses SAM3D to estimate object geometry and pose, and refines the pose through render-and-compare optimization. Subsequently, we render PartField descriptors from the reconstructed geometry into the image plane based on the estimated object pose. The resulting geometry-aware feature maps complement DINO and Stable Diffusion features, while geodesic distances on the reconstructed shapes enable reliable filtering of candidate correspondences. We use the filtered matches as supervision to train a lightweight adapter on top of DINO and Stable Diffusion for semantic correspondence. In contrast to prior post-training approaches that require pose annotations and rely on coarse spherical geometry, our method automatically obtains instance-specific 3D structure and uses it to guide correspondence learning. Experiments show that our approach improves semantic correspondence over the prior methods while reducing manual geometric supervision. Code and model can be found at https:/github.com/GenIntel/3D-SC.

Abstract

Semantic Correspondence

Explore similar work

Geometry Matters: 3D Foundation Priors for Learning Semantic Correspondence

Shape-of-You: Fused Gromov-Wasserstein Optimal Transport for Semantic Correspondence in-the-Wild

GeoWorld: Providing Full-frame Geometry Features to Facilitate 3D Scene Generation