# Fast Uncertainty Quantification for Deep Object Pose Estimation

arXiv: https://arxiv.org/abs/2011.07748 · HTML: https://mlsanity.com/paper/2011.07748/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Nov 2020.

## Built on by (most cited first, 18 of 18)

- [Look Closer: Bridging Egocentric and Third-Person Views with Transformers for Robotic Manipulation](https://mlsanity.com/paper/2201.07779/) (2022)
- [Challenges for Monocular 6D Object Pose Estimation in Robotics](https://mlsanity.com/paper/2307.12172/) (2023)
- [SpyroPose: SE(3) Pyramids for Object Pose Distribution Estimation](https://mlsanity.com/paper/2303.05308/) (2023)
- [Uncertainty Quantification with Deep Ensembles for 6D Object Pose Estimation](https://mlsanity.com/paper/2403.07741/) (2024)
- [Ki-Pode: Keypoint-based Implicit Pose Distribution Estimation of Rigid Objects](https://mlsanity.com/paper/2209.09659/) (2022)
- [Ambiguity-Aware Multi-Object Pose Optimization for Visually-Assisted Robot Manipulation](https://mlsanity.com/paper/2211.00960/) (2022)
- [Shape Completion with Prediction of Uncertain Regions](https://mlsanity.com/paper/2308.00377/) (2023)
- [DTF-Net: Category-Level Pose Estimation and Shape Reconstruction via Deformable Template Field](https://mlsanity.com/paper/2308.02239/) (2023)
- [Active 6D Pose Estimation for Textureless Objects using Multi-View RGB Frames](https://mlsanity.com/paper/2503.03726/) (2025)
- [Open Challenges for Monocular Single-shot 6D Object Pose Estimation](https://mlsanity.com/paper/2302.11827/) (2023)
- [Context-aware Multi-Model Object Detection for Diversely Heterogeneous Compute Systems](https://mlsanity.com/paper/2402.07415/) (2024)
- [Alignist: CAD-Informed Orientation Distribution Estimation by Fusing Shape and Correspondences](https://mlsanity.com/paper/2409.06683/) (2024)
- [MaskVal: Simple but Effective Uncertainty Quantification for 6D Pose Estimation](https://mlsanity.com/paper/2409.03556/) (2024)
- [Uncertainty Quantification for Visual Object Pose Estimation: S-Lemma Ellipsoidal Bounds](https://mlsanity.com/paper/2511.21666/) (2025)
- [COBRA -- COnfidence score Based on shape Regression Analysis for method-independent quality assessment of object pose estimation from single images](https://mlsanity.com/paper/2404.16471/) (2024)
- [Consensus-Driven Uncertainty for Robotic Grasping based on RGB Perception](https://mlsanity.com/paper/2506.20045/) (2025)
- [Deterministic Object Pose Confidence Region Estimation](https://mlsanity.com/paper/2506.22720/) (2025)
- [Detecting Pose Estimation Failures via Keypoint Self-Consistency](https://mlsanity.com/paper/2608.03516/) (2026)

## Abstract (arXiv)

Deep learning-based object pose estimators are often unreliable and overconfident especially when the input image is outside the training domain, for instance, with sim2real transfer. Efficient and robust uncertainty quantification (UQ) in pose estimators is critically needed in many robotic tasks. In this work, we propose a simple, efficient, and plug-and-play UQ method for 6-DoF object pose estimation. We ensemble 2-3 pre-trained models with different neural network architectures and/or training data sources, and compute their average pairwise disagreement against one another to obtain the uncertainty quantification. We propose four disagreement metrics, including a learned metric, and show that the average distance (ADD) is the best learning-free metric and it is only slightly worse than the learned metric, which requires labeled target data. Our method has several advantages compared to the prior art: 1) our method does not require any modification of the training process or the model inputs; and 2) it needs only one forward pass for each model. We evaluate the proposed UQ method on three tasks where our uncertainty quantification yields much stronger correlations with pose estimation errors than the baselines. Moreover, in a real robot grasping task, our method increases the grasping success rate from 35% to 90%.