noeyislearning's picture
docs: update metadata and dataset information
ed97924 verified
metadata
license: ecl-2.0
task_categories:
  - image-classification
language:
  - en
tags:
  - public
  - images
  - annotated
  - education
  - classification
pretty_name: '512x384 Litter Sort Annotated Wastes '
size_categories:
  - 1K<n<10K

(512x384) Litter Sort Annotated Wastes

It contains 2,527 annotated photographs of everyday waste, split into six material classes: Cardboard (403), Glass (501), Metal (410), Paper (594), Plastic (482) and miscellaneous Trash (137). Images were gathered from varied lighting conditions, backgrounds and device cameras to reflect realistic disposal scenes. Each file is placed in a folder named after its class, making the collection ready for direct ingestion into image-classification pipelines aimed at automated garbage-type detection.