Strength
What it is useful for
- Reduces rankings that depend on an arbitrary threshold choice.
- Can show both overall accuracy and performance in the low-error region.
6DoF / threshold aggregation · aggregate
ADD / ADD-S AUC
Area Under the ADD Recall Curve
Averages recall over a range of error thresholds instead of relying on a single threshold.
AUC integrates the curve of success rate versus error threshold. It is more continuous than ADD-0.1d, but the upper limit can substantially change the score.
Principle
What error does this metric penalize?
The formulas define the computation. The diagrams show whether the error concerns a 3D model, visible surfaces, the image plane, or aggregation across thresholds.
Each threshold θ corresponds to the fraction of instances with error below θ.
AUC averages over θ, distinguishing models near a threshold more effectively than a single threshold score.
How to compute
Computation steps
- 01
Collect the ADD or ADD-S error for every instance.
- 02
Compute recall at multiple thresholds between 0 and e_max.
- 03
Integrate numerically and normalize by e_max.
Caveat
Cautions when reading results
- Report the underlying error metric, object set, and AUC upper limit together.
- A high AUC does not guarantee strong performance at very strict thresholds.
Paper atlas
Which papers report this metric?
Links are derived by normalizing the metric names used in each paper. Directly reported metrics are distinguished from metrics associated through a broader protocol.
30Related papers
30Directly reported
0Linked through a protocol
2026 Directly reported
ConceptPose: Training-Free Zero-Shot Object Pose Estimation using Concept Vectors
CVPR · Liming Kuang, Yordanka Velikova, Mahdi Saleh et al.
Original terminology: ADD-AUC / ADD-S-AUC
2026 Directly reported
CoordAR: One-Reference 6D Pose Estimation of Novel Objects via Autoregressive Coordinate Map Generation
AAAI · Dexin Zuo, Ang Li, Wei Wang et al.
Original terminology: AUC ADD / AUC ADD-S
2026 Directly reported
Cov2Pose: Leveraging Spatial Covariance for Direct Manifold-aware 6-DoF Object Pose Estimation
CVPR · Nassim Ali Ousalah, Peyman Rostami, Vincent Gaudillière et al.
Original terminology: AUC ADD-S / AUC ADD(-S)
2026 Directly reported
EgoXtreme: A Dataset for Robust Object Pose Estimation in Egocentric Views under Extreme Conditions
CVPR · Taegyoon Yoon, Yegyu Han, Seojin Ji et al.
Original terminology: ADD-S AUC
2026 Directly reported
GenHOI: Generalized Hand-Object Pose Estimation with Occlusion Awareness
ECCV · Hui Yang, Wei Sun, Jian Liu et al.
Original terminology: ADD / ADD-S AUC
2026 Directly reported
GraspClutter6D: A Large-scale Real-world Dataset for Robust Perception and Grasping in Cluttered Scenes
ICRA / IEEE RA-L · Seunghyeok Back, Joosoon Lee, Kangmin Kim et al.
Original terminology: ADD(-S) and ADD-S AUC
2026 Directly reported
HIPPo: Harnessing Image-to-3D Priors for Model-Free Zero-Shot 6D Pose Estimation
ICRA / IEEE RA-L · Yibo Liu, Zhaodong Jiang, Binbin Xu et al.
Original terminology: ADD AUC / ADD-S AUC
2026 Directly reported
Deep Learning-Based Object Pose Estimation: A Comprehensive Survey
IJCV · Jian Liu, Wei Sun, Hui Yang et al.
Original terminology: ADD / ADD-S AUC / ADD AUC
2026 Directly reported
Pose Anything Anywhere: Model-free Object Poses from Arbitrary References
ECCV · Hongli Xu, Jiaqi Hu, Junwen Huang et al.
Original terminology: ADD / ADD-S AUC / ADD AUC
2026 Directly reported
Point2Pose: Occlusion-Recovering 6D Pose Tracking and 3D Reconstruction for Multiple Unknown Objects Via 2D Point Trackers
ECCV · Tzu-Yuan Lin, Ho Jae Lee, Kevin Doherty et al.
Original terminology: ADD-S AUC / ADD AUC
2025 Directly reported
Any6D: Model-free 6D Pose Estimation of Novel Objects
CVPR · Taeyeop Lee, Bowen Wen, Minjun Kang et al.
Original terminology: AUC ADD / AUC ADD-S
2025 Directly reported
BoxDreamer: Dreaming Box Corners for Generalizable Object Pose Estimation
ICCV · Yuanhong Yu, Xingyi He, Chen Zhao et al.
Original terminology: ADD-AUC / ADD-S AUC
2025 Directly reported
CRISP: Object Pose and Shape Estimation with Test-Time Adaptation
CVPR · Jingnan Shi, Rajat Talak, Harry Zhang et al.
Original terminology: ADD-S AUC
2025 Directly reported
DynamicPose: Real-time and Robust 6D Object Pose Tracking for Fast-Moving Cameras and Objects
IROS · Tingbang Liang, Yixin Zeng, Jiatong Xie et al.
Original terminology: ADD AUC at 0.1 m / ADD-S AUC at 0.1 m
2025 Directly reported
Occlusion-Aware 6D Pose Estimation with Depth-Guided Graph Encoding and Cross-Semantic Fusion for Robotic Grasping
ICRA · Jingyang Liu, Zhenyu Lu, Lu Chen et al.
Original terminology: ADD(-S) AUC
2025 Directly reported
ONDA-Pose: Occlusion-Aware Neural Domain Adaptation for Self-Supervised 6D Object Pose Estimation
CVPR · Tao Tan, Qiulei Dong
Original terminology: ADD-S AUC
2025 Directly reported
One2Any: One-Reference 6D Pose Estimation for Any Object
CVPR · Mengya Liu, Siyuan Li, Ajad Chhatkuli et al.
Original terminology: ADD-S AUC / ADD AUC
2025 Directly reported
SCFlow2: Plug-and-Play Object Pose Refiner with Shape-Constraint Scene Flow
CVPR · Qingyuan Wang, Rui Song, Jiaojiao Li et al.
Original terminology: ADD-S AUC
2025 Directly reported
UA-Pose: Uncertainty-Aware 6D Object Pose Estimation and Online Object Completion with Partial References
CVPR · Ming-Feng Li, Xin Yang, Fu-En Wang et al.
Original terminology: ADD-S AUC / ADD AUC
2025 Directly reported
ViTa-Zero: Zero-Shot Visuotactile Object 6D Pose Estimation
ICRA · Hongyu Li, James Akl, Srinath Sridhar et al.
Original terminology: ADD-S AUC / ADD AUC
2024 Directly reported
6D-Diff: A Keypoint Diffusion Framework for 6D Object Pose Estimation
CVPR · Li Xu, Haoxuan Qu, Yujun Cai et al.
Original terminology: ADD / ADD-S AUC
2024 Directly reported
FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects
CVPR · Bowen Wen, Wei Yang, Jan Kautz et al.
Original terminology: ADD / ADD-S AUC / ADD AUC / ADD-S AUC
2024 Directly reported
Learning Better Keypoints for Multi-Object 6DoF Pose Estimation
WACV · Yangzheng Wu, Michael Greenspan
Original terminology: ADD / ADD-S AUC / ADD(S) AUC
2024 Directly reported
Effects of Markers in Training Datasets on the Accuracy of 6D Pose Estimation
WACV · Janis Rosskamp, Rene Weller, Gabriel Zachmann
Original terminology: ADD / ADD-S AUC
2024 Directly reported
MOTPose: Multi-Object 6D Pose Estimation for Dynamic Video Sequences Using Attention-Based Temporal Fusion
ICRA · Arul Selvam Periyasamy, Sven Behnke
Original terminology: ADD AUC / ADD-S AUC / ADD(-S) AUC
2024 Directly reported
PACE: A Large-Scale Dataset with Pose Annotations in Cluttered Environments
ECCV · Yang You, Kai Xiong, Zhening Yang et al.
Original terminology: ADD AUC / ADD-S AUC
2024 Directly reported
Pseudo-keypoint RKHS Learning for Self-supervised 6DoF Pose Estimation
ECCV · Yangzheng Wu, Michael Greenspan
Original terminology: ADD(S) AUC / ADD-S AUC
2024 Directly reported
YOLO-6D-Pose: Enhancing YOLO for Single-Stage Monocular Multi-Object 6D Pose Estimation
3DV · Debapriya Maji, Soyeb N. Nagori, Manu Mathew et al.
Original terminology: ADD(-S) AUC
2022 Directly reported
Gen6D: Generalizable Model-Free 6-DoF Object Pose Estimation from RGB Images
ECCV · Yuan Liu, Yilin Wen, Sida Peng et al.
Original terminology: ADD-AUC
2022 Directly reported
PIZZA: A Powerful Image-only Zero-Shot Zero-CAD Approach to 6 DoF Tracking
3DV · Van Nguyen Nguyen, Yuming Du, Yang Xiao et al.
Original terminology: ADD-S AUC
Next
Compare related metrics
Compare metrics at the same level to understand why one score improves while another stays unchanged.
ADD
Average Distance of Model Points
error
Compares each CAD point under predicted and ground-truth poses, averaging the 3D distances.
ADD-S
Average Closest Point Distance
error
Relaxes point identity: each ground-truth point is matched to the nearest point in the predicted model.
ADD(S) / ADD(-S)
Symmetry-dependent ADD Reporting Protocol
protocol
Selects ADD or ADD-S according to object symmetry, then aggregates the results into one score.