A Multi-Level Approach to Waste Object Segmentation

TL;DR

This paper presents a multi-level segmentation approach combining deep learning and CRFs to accurately localize waste objects in RGB-D images, supported by a new dataset and validated on multiple benchmarks.

Contribution

The paper introduces a novel multi-level segmentation framework for waste objects using RGB-D data and provides a new public dataset for this task.

Findings

Effective segmentation on MJU-Waste and TACO datasets

Improved accuracy over baseline methods

Public release of the MJU-Waste dataset

Abstract

We address the problem of localizing waste objects from a color image and an optional depth image, which is a key perception component for robotic interaction with such objects. Specifically, our method integrates the intensity and depth information at multiple levels of spatial granularity. Firstly, a scene-level deep network produces an initial coarse segmentation, based on which we select a few potential object regions to zoom in and perform fine segmentation. The results of the above steps are further integrated into a densely connected conditional random field that learns to respect the appearance, depth, and spatial affinities with pixel-level accuracy. In addition, we create a new RGBD waste object segmentation dataset, MJU-Waste, that is made public to facilitate future research in this area. The efficacy of our method is validated on both MJU-Waste and the Trash Annotation in Context (TACO) dataset.