Incremental Object 6D Pose Estimation

Long Tian, Amelia Sorrenti, Yik Lung Pang, Giovanni Bellitto, Simone Palazzo, Concetto Spampinato, Changjae Oh

Abstract: We present a novel setting for 6D object pose estimation, where a model progressively adapts its parameters to estimate the pose of new objects without forgetting. This capability is crucial for real-world applications, particularly in scenarios where a deployed model must accommodate new objects while mitigating the risk of forgetting previously seen objects. To tackle this challenge, we propose a replay-based incremental learning technique designed to retain key information about previously seen objects when the model is exposed to a new one. Our approach relies on a memory buffer comprising keyframes of previously encountered objects, serving to regularize the model parameters based on past experiences while allowing for the update of model features to perform pose estimation on new objects. We validate the effectiveness of our method on the standard Linemod and YCB-Video datasets, demonstrating how our method surpasses baseline approaches in incremental learning at the task at hand. The project website is available at: https://qm-ipalab.github.io/ILPose.