6DoF / threshold accuracy · threshold

n° mcm

Rotation–Translation Threshold Accuracy

Counts a pose as correct only when both rotation and translation errors fall below their thresholds.

Examples include 5° 2cm, 5° 5cm, and 10° 5cm. Common in category-level and 9DoF papers, but results at different thresholds are not directly comparable.

Explore principles and formulas View related papers 32

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.

Converts continuous errors into a binary decision about whether the required precision is met.

The joint criterion prevents a correct rotation alone, or a correct position alone, from being counted as a successful pose.

\operatorname{Acc}_{n^\circ,m\mathrm{cm}} = \frac{1}{N}\sum_{i=1}^{N}\mathbf{1}[e_R^{(i)} < n^\circ \;\wedge\; e_t^{(i)} < m\mathrm{cm}]

Thresholds are part of the protocol; ranking 5°2cm and 10°5cm scores together is meaningless.

Illustration created for this site to explain the metric's error space; it does not show measured data or a figure from the original paper.

How to compute

Computation steps

01
Compute rotation and translation errors for each instance.
02
Determine success using the n° and mcm thresholds specified by the paper or dataset.
03
Compute the success fraction over all instances; report each threshold in a separate column when using multiple thresholds.

Strength

What it is useful for

Directly reflects application tolerances.
Independent of CAD surface sampling and suitable for category-level scale and pose tasks.

Caveat

Cautions when reading results

Binarization can conceal continuous improvements near a threshold.
A fixed cm threshold is not always fair across objects of different sizes.

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.

Filter the paper library

32Related papers

32Directly reported

0Linked through a protocol

2026 Directly reported

ComPose: A Unified Completion-Pose Framework for Robust Category-Level Object Pose Estimation

CVPR · Huan Ren, Yihan Chen, Chuxin Wang et al.

Original terminology: 10 degree 2 cm accuracy

2026 Directly reported

MemPose: Category-level Object Pose Estimation with Memory

ECCV · Xiao Lin, Minghao Zhu, Yun Peng et al.

Original terminology: n° mcm / 5° 2cm / 5° 5cm / 10° 2cm / 10° 5cm

2026 Directly reported

Deep Learning-Based Object Pose Estimation: A Comprehensive Survey

IJCV · Jian Liu, Wei Sun, Hui Yang et al.

n° mcm

What error does this metric penalize?

Computation steps

What it is useful for

Cautions when reading results

Which papers report this metric?

ComPose: A Unified Completion-Pose Framework for Robust Category-Level Object Pose Estimation

MemPose: Category-level Object Pose Estimation with Memory

Deep Learning-Based Object Pose Estimation: A Comprehensive Survey

CAP-Net: A Unified Network for 6D Pose and Size Estimation of Categorical Articulated Parts from a Single RGB-D Image

CleanPose: Category-Level Object Pose Estimation via Causal Learning and Knowledge Distillation

GCE-Pose: Global Context Enhancement for Category-level Object Pose Estimation

Generalizable Single-View Object Pose Estimation by Two-Side Generating and Matching

GIVEPose: Gradual Intra-class Variation Elimination for RGB-based Category-Level Object Pose Estimation

Leveraging Global Stereo Consistency for Category-Level Shape and 6D Pose Estimation from Stereo Images

KeyPose: Category-Level 6D Object Pose Estimation with Self-Adaptive Keypoints

MK-Pose: Category-Level Object Pose Estimation via Multimodal-Based Keypoint Learning

RCGNet: RGB-based Category-Level 6D Object Pose Estimation with Geometric Guidance

Rethinking Correspondence-based Category-Level Object Pose Estimation

Universal Features Guided Zero-Shot Category-Level Object Pose Estimation

Instance-Adaptive and Geometric-Aware Keypoint Learning for Category-Level 6D Object Pose Estimation

DiffusionNOCS: Managing Symmetry and Uncertainty in Sim2Real Multi-Modal Category-Level Pose Estimation

RGB-Based Category-Level Object Pose Estimation via Decoupled Metric Scale Recovery

Implicit Coarse-to-Fine 3D Perception for Category-Level Object Pose Estimation from Monocular RGB Image

MFOS: Model-Free & One-Shot Object Pose Estimation

MV-ROPE: Multi-View Constraints for Robust Category-Level Object Pose and Size Estimation

Marrying NeRF with Feature Matching for One-Step Pose Estimation

NeuSurfEmb: A Complete Pipeline for Dense Correspondence-Based 6D Object Pose Estimation without CAD Models

OP-Align: Object-level and Part-level Alignment for Self-supervised Category-level Articulated Object Pose Estimation

Robust Category-Level 3D Pose Estimation From Diffusion-Enhanced Synthetic Data

Scalable Network and Adaptive Refinement Module for 6D Pose Estimation of Diverse Industrial Components

Single-Shot 6DoF Pose and 3D Size Estimation for Robotic Strawberry Harvesting

ToolEENet: Tool Affordance 6D Pose Estimation

Zero123-6D: Zero-Shot Novel View Synthesis for RGB Category-Level 6D Pose Estimation

CenterSnap: Single-Shot Multi-Object 3D Shape Reconstruction and Categorical 6D Pose and Size Estimation

MegaPose: 6D Pose Estimation of Novel Objects via Render & Compare

PIZZA: A Powerful Image-only Zero-Shot Zero-CAD Approach to 6 DoF Tracking

Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimation

Compare related metrics

Rotation Error

Translation Error

3D IoU