Keypoint-Based Category-Level Object Pose Tracking from an RGB Sequence with Uncertainty Estimation

TL;DR

This paper introduces a probabilistic, single-stage method for category-level 6-DoF object pose tracking from RGB video that leverages previous frame predictions and uncertainties for improved accuracy and stability.

Contribution

It presents a novel probabilistic filtering approach that incorporates previous uncertainties into pose estimation, enhancing robustness over existing single-frame methods.

Findings

Outperforms existing methods on Objectron benchmark

Provides more accurate and stable pose predictions

Demonstrates effectiveness in augmented reality applications

Abstract

We propose a single-stage, category-level 6-DoF pose estimation algorithm that simultaneously detects and tracks instances of objects within a known category. Our method takes as input the previous and current frame from a monocular RGB video, as well as predictions from the previous frame, to predict the bounding cuboid and 6-DoF pose (up to scale). Internally, a deep network predicts distributions over object keypoints (vertices of the bounding cuboid) in image coordinates, after which a novel probabilistic filtering process integrates across estimates before computing the final pose using PnP. Our framework allows the system to take previous uncertainties into consideration when predicting the current frame, resulting in predictions that are more accurate and stable than single frame methods. Extensive experiments show that our method outperforms existing approaches on the challenging Objectron benchmark of annotated object videos. We also demonstrate the usability of our work in an augmented reality setting.