Fast Uncertainty Quantification for Deep Object Pose Estimation

by   Guanya Shi, et al.

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


page 1

page 5

page 6


Deep Directional Statistics: Pose Estimation with Uncertainty Quantification

Modern deep learning systems successfully solve many perception tasks su...

Investigations on Output Parameterizations of Neural Networks for Single Shot 6D Object Pose Estimation

Single shot approaches have demonstrated tremendous success on various c...

Sample-based Uncertainty Quantification with a Single Deterministic Neural Network

Development of an accurate, flexible, and numerically efficient uncertai...

Post-hoc Uncertainty Learning using a Dirichlet Meta-Model

It is known that neural networks have the problem of being over-confiden...

Object-Driven Active Mapping for More Accurate Object Pose Estimation and Robotic Grasping

This paper presents the first active object mapping framework for comple...

SLAM-Supported Self-Training for 6D Object Pose Estimation

Recent progress in learning-based object pose estimation paves the way f...

Data-driven Prognostics with Predictive Uncertainty Estimation using Ensemble of Deep Ordinal Regression Models

Prognostics or Remaining Useful Life (RUL) Estimation from multi-sensor ...

Please sign up or login with your details

Forgot password? Click here to reset