Published June 4, 2021 | Version 1.2.1
Dataset Open

ZeroWaste Dataset: Towards Deformable Object Segmentation in Cluttered Scenes

  • 1. Boston University
  • 2. The American University in Cairo
  • 3. Worcester Polytechnic Institute
  • 4. University of Washington
  • 5. Boston University and MIT-IBM Watson AI Lab


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.   Recyclable waste detection poses a unique computer vision challenge as it requires detection of highly deformable and often translucent objects in cluttered scenes without the kind of context information usually present in human-centric datasets. 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. 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

Please use the following password to extract the zip files: UP#1VuX409z4


Files (30.0 GB)

Name Size Download all
2.1 GB Download
2.1 GB Download
2.1 GB Download
2.1 GB Preview Download
7.5 GB Preview Download
3.2 GB Download
3.2 GB Download
3.2 GB Download
966.3 MB Preview Download
3.3 GB Preview Download