UDAPose: Unsupervised Domain Adaptation for Low-Light Human Pose Estimation

Haopeng Chen; Yihao Ai; Kabeen Kim; Robby T. Tan; Yixin Chen; Bo Wang

cs.CVApr 12, 2026

UDAPose: Unsupervised Domain Adaptation for Low-Light Human Pose Estimation

Authors: Haopeng Chen, Yihao Ai, Kabeen Kim, Robby T. Tan, Yixin Chen, Bo Wang

Organizations: University of Mississippi · National University of Singapore · Duksung Women’s University · ASUS Intelligent Cloud Services (AICS)

Abstract

Low-visibility scenarios, such as low-light conditions, pose significant challenges to human pose estimation due to the scarcity of annotated low-light datasets and the loss of visual information under poor illumination. Recent domain adaptation techniques attempt to utilize well-lit labels by augmenting well-lit images to mimic low-light conditions. But handcrafted augmentations oversimplify noise patterns, while learning-based methods often fail to preserve high-frequency low-light characteristics, producing unrealistic images that lead pose models to generalize poorly to real low-light scenes. Moreover, recent pose estimators rely on image cues through image-to-keypoint cross-attention, but these cues become unreliable under low-light conditions. To address these issues, we propose Unsupervised Domain Adaptation for Pose Estimation (UDAPose), a novel framework that synthesizes low-light images and dynamically fuses visual cues with pose priors for improved pose estimation. Specifically, our synthesis method incorporates a Direct-Current-based High-Pass Filter (DHF) and a Low-light Characteristics Injection Module (LCIM) to inject high-frequency details from input low-light images, overcoming rigidity or the detail loss in existing approaches. Furthermore, we introduce a Dynamic Control of Attention (DCA) module that adaptively balances image cues with learned pose priors in the Transformer architecture. Experiments show that UDAPose outperforms state-of-the-art methods, with notable AP gains of 10.1 (56.4%) on the ExLPose-test hard set (LL-H) and 7.4 (31.4%) in cross-dataset validation on EHPT-XC. Code: https://github.com/Vision-and-Multimodal-Intelligence-Lab/UDAPose

Figures & tables

Panel of Figure 2 — Figure 2: Limitations of learning-based low-light augmentation. The first two columns show well-lit and paired low-light images from ExLPose [ 30 ] . The third and fourth columns present results from CycleGAN [ 80 ] and StyleID [ 9 ] . The last column shows our result. Low-light images are scaled to an average channel intensity of 0.4 for visualization only.

Figure 3: Overview of the UDAPose framework. During augmentation, the LCIM uses extracted low-light features from unpaired low-light images ( $I_{\text{LL}}$ ) to synthesize low-light counterparts ( $\hat{I}_{\text{LL}}$ ) of well-lit images ( $I_{\text{WL}}$ ). These synthetic images retain the original pose annotations from $I_{\text{WL}}$ while accurately reflecting low-light characteristics of $I_{\text{LL}}$ . The pose model is trained using these augmented images and their inherited annotations. Our DCA adaptively balances image cues and pose priors under low-light conditions. During inference, the trained model is directly applied to real low-light images. Note that $I_{\text{LL}}$ , $I^{\prime}_{\text{LL}}$ , and $\hat{I}_{\text{LL}}$ are scaled for visualization only.

Panel of Figure 4 — Figure 4: Ratio of Frobenius Norm of $\mathbf{Q}_{\text{image}}$ over $\mathbf{Q}_{\text{pose}}$ (i.e. $\|\mathbf{Q}_{\text{image}}\|_{2}/\|\mathbf{Q}_{\text{pose}}\|_{2}$ ) on different keypoints and pose estimation results before/after applying DCA module. Note that images are scaled for visualization only.

Methods	AP ↑ @.50:.95
Methods	WL	LL-A	LL-N	LL-H	LL-E
RFormer [ 3 ]	60.0	4.5	15.2	0.3	0.8
DarkIR [ 13 ]	60.2	6.1	17.4	1.3	1.2
LightenDiff [ 19 ]	60.1	5.6	13.9	0.7	0.8
QuadPrior [ 65 ]	60.2	8.9	19.3	4.6	0.3
CycleGAN [ 80 ]	61.3	19.6	33.7	17.9	3.3

Table 1: Evaluation mAP on ExLPose-test comparing image enhancement and domain adaptation methods. All methods are trained only on augmented images and well-lit annotations, without using low-light ground truth or paired data. The best is bold . The second best is underlined .

Methods	AR ↑ @.50:.95
Methods	WL	LL-A	LL-N	LL-H	LL-E
RFormer [ 3 ]	71.5	9.7	25.4	4.1	0.6
DarkIR [ 13 ]	72.0	11.1	29.3	7.3	0.8
LightenDiff [ 19 ]	71.8	10.2	22.9	4.9	0.5
QuadPrior [ 65 ]	71.9	15.5	30.9	11.7	1.0
CycleGAN [ 80 ]	72.1	28.7	45.6	27.4	8.9

Table 2: Evaluation mAR on ExLPose-test comparing image enhancement and domain adaptation methods, following Tab. 1 . The best is bold . The second best is underlined .

Panel of Figure 5 — Figure 5: Qualitative comparisons of our method with existing baselines, including image enhancement [ 65 ] , domain adaptation [ 1 ] , and image translation method [ 26 ] . From top to bottom, we show samples from low-light normal, hard, and extreme sets from ExLPose-test [ 30 ] . Our approach consistently outperforms the existing methods in human pose estimation across all scenarios. Images are scaled up for visualization purpose only.

Methods	AP ↑ @.50:.95			AR ↑ @.50:.95
Methods	Avg.	A7 M3	RIC OH3	Avg.	A7 M3	RIC OH3
RFormer [ 3 ]	27.5	29.4	25.7	43.7	47.2	40.3
DarkIR [ 13 ]	28.9	30.3	27.6	47.0	48.7	45.4
LightenDiff [ 19 ]	25.3	29.1	21.5	40.6	45.4	35.9
QuadPrior [ 65 ]	29.3	30.6	28.0	48.8	50.4	47.2
CycleGAN [ 80 ]	45.1	47.5	42.7	60.7	63.7	57.8

Table 3: Evaluation on ExLPose-OCN, following identical setup as in Tab. 1 . The best is bold . The second best is underlined .

Methods	AP ↑			AR ↑
Methods	@.50:.95	@.50	@.75	@.50:.95	@.50	@.75
RFormer [ 3 ]	8.8	20.3	7.9	21.3	39.2	18.9
DarkIR [ 13 ]	12.5	25.2	11.3	28.9	48.5	27.1
LightenDiff [ 19 ]	9.7	17.7	8.3	22.0	38.4	20.3
QuadPrior [ 65 ]	12.9	24.5	11.5	28.1	47.7	27.5
CycleGAN [ 80 ]	20.7	36.9	19.2	45.0	69.6	45.5

Table 4: Cross-dataset validation on EHPT-XC [ 8 ] , using the model weights as in Tab. 1 . Best is bold , second best underlined .

	AP ↑ @0.5:0.95
	WL	LL-N	LL-H	LL-E	A7 M3	RIC OH3	EHPT -XC
Well-lit	60.1	3.4	0.4	0.2	11.3	13.4	0.3
HM	60.0	21.3	1.3	0.5	15.1	11.4	1.5
Baseline SD	60.0	23.7	7.2	0.0	30.1	25.4	3.2
+ AIN	60.1	25.2	8.8	2.4	33.0	27.8	6.1
+ LCIM	60.1	31.5	20.7	7.8	43.1	39.8	19.5

Table 5: Ablation study of our proposed modules on ExLPose-test, ExLPose-OCN, and EHPT-XC. Well-lit: pose model trained with well-lit images only. HM: pose model adapted with synthetic low-light images using histogram matching. AIN is a normalization step (see supplementary). The best is bold .

	AP ↑ @0.5:0.95
	WL	LL-N	LL-H	LL-E	A7 M3	RIC OH3	EHPT -XC
Ours (ExLPose)	67.3	38.7	28.0	11.7	55.0	47.9	31.0
Ours (CrowdPose)	71.7	47.9	37.2	19.4	70.5	69.7	45.7

Table 6: Comparison results of using ExLPose well-lit and CrowdPose to construct synthetic low-light training data.

Figure 6: The architecture of our DCA module.

	AP ↑ @0.5:0.95
	WL	LL-N	LL-H	LL-E	A7 M3	RIC OH3	EHPT -XC
4,000	66.1	34.7	22.4	5.4	50.0	45.1	25.4
8,000	66.3	35.4	24.8	7.8	53.2	46.8	27.9
12,000	66.9	36.6	26.2	10.4	54.7	47.2	29.1
16,000	66.8	37.5	27.3	11.3	55.0	47.8	30.5
20,000	67.3	38.7	28.0	11.7	55.0	47.9	31.0

Table 7: Performance vs. amount of synthetic training data.

Methods	PSNR ↑	SSIM ↑	LPIPS ↓	FID ↓	KL ↓
CycleGAN [ 80 ]	36.56	0.76	0.26	50.20	0.028
UNIT [ 39 ]	33.90	0.66	0.29	45.70	0.104
UNSB [ 26 ]	34.19	0.74	0.30	96.42	0.062
EnCo [ 2 ]	35.99	0.75	0.28	48.71	0.031
Ours	41.13	0.91	0.20	11.17	0.008

Table 8: Evaluation for human pose anatomical consistency of our method and learning-based baselines.

	AP ↑ @0.5:0.95
	WL	LL-N	LL-H	LL-E	A7 M3	RIC OH3
Main (ELLA)	62.1	29.4	13.6	1.6	35.0	27.2
Main (Ours)	61.5	32.3	23.2	8.3	37.2	35.0
Comp. (ELLA)	60.3	27.8	11.9	0.8	33.9	26.5
Comp. (Ours)	60.8	31.7	22.4	6.7	36.8	33.9
Student (ELLA)	60.8	35.6	18.6	5.0	39.1	36.2

Table 9: Full comparison results of ELLA [ 1 ] and our method. “Main” refers to “main teacher”. “Comp.” refers to “complementary teacher”. And “student” refers to “student” distillation model, which is the full model of ELLA. The best is bold .

	AP ↑ @0.5:0.95
	WL	LL-N	LL-H	LL-E	A7 M3	RIC OH3
Direct input	60.8	23.7	7.3	0.0	27.3	24.3
Z-score-based norm.	60.9	25.4	13.2	2.2	28.4	25.0
Fixed factor	58.1	29.0	20.8	6.4	33.2	31.0
ImageNet-based	61.3	31.7	22.4	7.4	36.5	33.8
Ours	61.5	32.3	23.2	8.3	37.2	35.0

Table 10: Evaluation of the AIN module on ExLPose-test and ExLPose-OCN. Direct input refers to feeding low-light images into SD without AIN. Experiments are conducted using the DEKR pose model [ 14 ] , with DHF and LCIM enabled for all normalization approaches. The best is bold .

	AP ↑ @0.5:0.95
	WL	LL-N	LL-H	LL-E	A7 M3	RIC OH3	EHPT -XC
$z_{0}$	66.8	29.7	12.1	0.1	40.0	36.8	11.3
$z_{0},z_{1}$	66.8	31.4	19.2	2.4	41.4	37.4	15.4
$z_{0},z_{1},z_{2}$	67.2	35.3	23.2	5.8	43.8	39.9	26.2
$z_{0},z_{1},z_{2},z_{3}$	67.4	37.7	26.5	7.7	47.9	43.7	29.7
$z_{0},z_{1},z_{2},z_{3},z_{4}$	67.3	38.7	28.0	11.7	55.0	47.9	31.0

Table 11: Ablation study of LCIM on ExLPose-test, ExLPose-OCN, and EHPT-XC.

z_{0}

refers to baseline SD without any extra intermediate features.

z_{1}

z_{4}

represent low-to-high-frequency information fused in a coarse-to-fine integration strategy. Results are reported with AIN, DHF and DCA. The best is bold .

Panel of Figure 7 — Figure 7: (a) Effect of $\lambda$ . (b) Masking evaluation w/ and w/o DCA.

	AP ↑ @0.5:0.95
	WL	LL-N	LL-H	LL-E	A7 M3	RIC OH3	EHPT -XC
SE-Block [ 16 ]	62.4	36.7	26.3	9.5	50.3	46.5	26.7
CBAM [ 68 ]	62.5	37.0	26.2	9.8	51.1	46.2	27.0
Ours (DCA)	67.3	38.7	28.0	11.7	55.0	47.9	31.0

Table 12: Comparison of SE-Block [ 16 ] , CBAM [ 68 ] , and our DCA gating mechanism. The best is bold .

	AP ↑ @0.5:0.95
	WL	LL-N	LL-H	LL-E	A7 M3	RIC OH3	EHPT -XC
ControlNet [ 77 ]	66.3	31.7	16.4	2.7	47.6	43.7	22.4
IP-Adapter [ 76 ]	65.8	31.5	17.1	3.5	48.4	43.4	24.1
Ours	67.3	38.7	28.0	11.7	55.0	47.9	31.0

Table 13: Comparison of ControlNet [ 77 ] , IP-Adapter [ 76 ] , and our method. The best is bold .

Panel of Figure 8 — Figure 8: Qualitative ablation of our DCA module. L. represents left, R. represents right.

Panel of Figure 9 — Figure 9: Pose predictions of UDAPose compared against competing methods. The first two columns show results from enhancement-based methods; all other columns display results on the original low-light images. The low-light images are scaled for visualization only.

Panel of Figure 10 — Figure 10: Qualitative comparison of our data synthesis method with baselines.

Human Pose Estimation

Human pose estimation recovers people’s body keypoints, 3D skeletons, or meshes from images, video, and other sensor data.

See all 79 papers

Unsupervised Domain Adaptation for Sim-to-Real Object Pose Estimation with Contrastive Alignment and Pseudo-Label Refinement
Jun 19, 2026Nidhal Eddine Chenni, Arunkumar Rathinam, Djamila AouadaSynthetic-to-Real Domain Adaptation Unsupervised Domain Adaptation
SnT, University of Luxembourg
SDPose: Exploiting Diffusion Priors for Out-of-Domain and Robust Pose Estimation
Sep 29, 2025Shuang Liang, Jing He, Lejun Liao +4Human Pose Estimation Visual Representation Learning
Rama Alpaca Technology · The University of Hong Kong · HKUST (GZ) +3
Dark3R: Learning Structure from Motion in the Dark
Mar 5, 2026Andrew Y Guo, Anagh Malik, SaiKiran Tedla +7Structure from Motion Camera Pose Estimation
University of Toronto · Vector Institute · York University +3

Jun 19, 2026J/K move · Enter open · S save

Unsupervised Domain Adaptation for Sim-to-Real Object Pose Estimation with Contrastive Alignment and Pseudo-Label Refinement

Nidhal Eddine Chenni, Arunkumar Rathinam, Djamila Aouada

SnT, University of Luxembourg

Synthetic-to-Real Domain Adaptation Unsupervised Domain Adaptation Object Pose Estimation

Open paper AskPDF arXiv

Unsupervised domain adaptation (UDA) enables robust transfer of knowledge from simulated to real environments while exploiting a subset of unlabeled target data to improve real-world performance. Existing UDA methods for Object pose estimation often rely on global feature matching, multi-stage larger frameworks, or image translation pipelines, which tend to overlook the pose-specific information embedded in feature representations. To bridge this limitation, we introduce CAPLR that targets the adaptation of pose-sensitive features in localized regions, ensuring that domain alignment preserves the geometric cues essential for accurate pose estimation. CAPLR achieves UDA with three key components: (1) Efficient Cross-Domain Pairing strategy leveraging intermediate features to identify pose similar image pairs across domains without supervision; (2) Contrastive Alignment to perform feature alignment at localised regions in both intermediate and task-specific representations; and (3) Consistency-Based Pseudo-Label Refinement to improve reliability by encouraging stable target predictions. Extensive experiments demonstrate that CAPLR achieves state-of-the-art performance across multiple well-known object pose estimation benchmarks featuring diverse and challenging scenarios.

Abstract

Human Pose Estimation

Explore similar work

Unsupervised Domain Adaptation for Sim-to-Real Object Pose Estimation with Contrastive Alignment and Pseudo-Label Refinement

SDPose: Exploiting Diffusion Priors for Out-of-Domain and Robust Pose Estimation

Dark3R: Learning Structure from Motion in the Dark