TACO: Trash Annotations in Context for Litter Detection

TL;DR

TACO is a growing open dataset for litter detection and segmentation, utilizing crowdsourcing, with initial promising results using Mask R-CNN, but requiring more annotations for real-world deployment.

Contribution

This paper introduces the TACO dataset for litter detection, along with tools and initial segmentation performance results, highlighting the need for further annotations.

Findings

Mask R-CNN achieves promising segmentation results on TACO

TACO dataset contains 1500 images and 4784 annotations

Further manual annotations are needed for real-world application

Abstract

TACO is an open image dataset for litter detection and segmentation, which is growing through crowdsourcing. Firstly, this paper describes this dataset and the tools developed to support it. Secondly, we report instance segmentation performance using Mask R-CNN on the current version of TACO. Despite its small size (1500 images and 4784 annotations), our results are promising on this challenging problem. However, to achieve satisfactory trash detection in the wild for deployment, TACO still needs much more manual annotations. These can be contributed using: http://tacodataset.org/