SuperPose: Improved 6D Pose Estimation with Robust Tracking and Mask-Free Initialization

TL;DR

SuperPose offers a real-time, mask-free 6D object detection method that combines FoundationPose, SAM2, and LightGlue, effectively handling symmetric objects and tracking challenges without retraining, suitable for industrial applications.

Contribution

The paper introduces a novel 6D pose estimation approach that eliminates the need for initial masks and retraining, integrating multiple models for robust, real-time industrial object tracking.

Findings

Reliable 6D detection on YCB dataset and industrial objects

Effective handling of occlusion and rapid movement

Continuous tracking without retraining or initial masks

Abstract

We developed a robust solution for real-time 6D object detection in industrial applications by integrating FoundationPose, SAM2, and LightGlue, eliminating the need for retraining. Our approach addresses two key challenges: the requirement for an initial object mask in the first frame in FoundationPose and issues with tracking loss and automatic rotation for symmetric objects. The algorithm requires only a CAD model of the target object, with the user clicking on its location in the live feed during the initial setup. Once set, the algorithm automatically saves a reference image of the object and, in subsequent runs, employs LightGlue for feature matching between the object and the real-time scene, providing an initial prompt for detection. Tested on the YCB dataset and industrial components such as bleach cleanser and gears, the algorithm demonstrated reliable 6D detection and tracking. By integrating SAM2 and FoundationPose, we effectively mitigated common limitations such as the problem of tracking loss, ensuring continuous and accurate tracking under challenging conditions like occlusion or rapid movement.