ImageNet 1K Wds: Image Dataset Organized by WordNet Synsets

ILSVRC 2012, commonly known as 'ImageNet', organizes images according to the WordNet hierarchy. Each meaningful concept, or 'synset', is illustrated with an average of 1000 quality-controlled, human-annotated images. The dataset was uploaded to Hugging Face by 'timm' and last updated on January 7, 2024.

Use Cases

• Training image classification models based on the hierarchical WordNet synset structure.
• Benchmarking object recognition algorithms based on the quality-controlled, human-annotated images.
• Developing multi-label classification systems based on the synonym set (synset) organization.
• Studying visual concept representation based on the link between images and linguistic concepts in WordNet.

Strengths

• Images are organized according to the established WordNet hierarchy with over 100,000 synsets.
• Each synset is illustrated with an average of 1000 images, providing substantial visual examples per concept.
• All images are quality-controlled and human-annotated, suggesting a high standard for label accuracy.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and dataset size are unknown, which may limit suitability assessment.
• The license is unknown, which could restrict commercial or research use.

Provenance

Source

Hugging Face dataset uploaded by author 'timm'.

Collection Method

Images were quality-controlled and human-annotated to illustrate WordNet synsets.

Time Range

Associated with the ILSVRC 2012 challenge.

Freshness

Last updated 2024-01-07 18:12:43; freshness should be verified.

Geography

null