Active Perception for Ambiguous Objects Classification

TL;DR

This paper introduces an active perception framework that iteratively selects viewpoints to resolve object ambiguities, improving classification accuracy for ambiguous objects in real-world scenarios.

Contribution

It presents a novel active perception method that guides viewpoint selection to disambiguate objects from a single view, with a complete pipeline and real-world validation.

Findings

Effective viewpoint selection reduces classification ambiguities.

Validated on household objects with a robotic platform.

Source code released for reproducibility.

Abstract

Recent visual pose estimation and tracking solutions provide notable results on popular datasets such as T-LESS and YCB. However, in the real world, we can find ambiguous objects that do not allow exact classification and detection from a single view. In this work, we propose a framework that, given a single view of an object, provides the coordinates of a next viewpoint to discriminate the object against similar ones, if any, and eliminates ambiguities. We also describe a complete pipeline from a real object's scans to the viewpoint selection and classification. We validate our approach with a Franka Emika Panda robot and common household objects featured with ambiguities. We released the source code to reproduce our experiments.