ZeroWaste Dataset: Towards Deformable Object Segmentation in Extreme Clutter

Less than 35% of recyclable waste is being actually recycled in the US, which leads to increased soil and sea pollution and is one of the major concerns of environmental researchers as well as the common public. At the heart of the problem are the inefficiencies of the waste sorting process (separating paper, plastic, metal, glass, etc.) due to the extremely complex and cluttered nature of the waste stream. Automated waste detection has great potential to enable more efficient, reliable, and safe waste sorting practices, but it requires label-efficient detection of deformable objects in extremely cluttered scenes. This challenging computer vision task currently lacks suitable datasets or methods in the available literature. In this paper, we take a step towards computer-aided waste detection and present the first in-the-wild industrial-grade waste detection and segmentation dataset, ZeroWaste. This dataset contains over 1800 fully segmented video frames collected from a real waste sorting plant along with waste material labels for training and evaluation of the segmentation methods, as well as over 6000 unlabeled frames that can be further used for semi-supervised and self-supervised learning techniques, as well as frames of the conveyor belt before and after the sorting process, comprising a novel setup that can be used for weakly-supervised segmentation. Our experimental results demonstrate that state-of-the-art segmentation methods struggle to correctly detect and classify target objects which suggests the challenging nature of our proposed real-world task of fine-grained object detection in cluttered scenes. We believe that ZeroWaste will catalyze research in object detection and semantic segmentation in extreme clutter as well as applications in the recycling domain. Our project page can be found at http://ai.bu.edu/zerowaste/.