Oriel — AI & technology worth noticing, every day.

timmzak/ReText-Captions

Dataset · https://huggingface.co/datasets/timmzak/ReText-Captions

ReText: Generated Captions for Multi-Camera Person Re-identification Accepted by WACV 2027 Paper · GitHub · VLM weights Overview Generated textual annotations for CUHK03, Market-1501, MSMT17, CUHK-SYSU, and RandPerson, released with ReText. The captions were generated using LLaVA fine-tuned on UFine3C. This release contains image paths and captions; original images are not included. Contents Each dataset is available as a separate… See the full description on the dataset page: https://huggingface.co/datasets/timmzak/ReText-Captions.

First seen by Oriel on .

AI summary

The Hugging Face dataset contains image paths paired with natural-language captions describing people’s clothing, accessories, appearance, and sometimes held objects.

  • Entries use paths such as `train/02534_s3119_0014785_0.jpg` followed by a caption.
  • Captions describe garments, colors, footwear, bags, hats, hairstyles, and accessories in detailed or concise language.
  • Most examples focus on people photographed in casual, formal, or everyday outfits.
  • The extracted content does not provide dataset size, licensing, splits beyond `train`, or collection methodology.

#image captioning #computer vision #fashion #datasets

Generated by gpt-5.6-sol. Check the original for details.

Metric trends

Last 30 days · Up to 120 recorded samples per source and metric, with reposts kept separate · First and latest retained

Likes

Total likes reported by the source.

  • Hugging Face Datasets: 12, latest observation ; +2 over observed period since

Downloads (30d)

Rolling 30-day downloads. A change in this value is not new downloads during the observed period.

  • Hugging Face Datasets: 248, latest observation ; +248 over observed period since

Aliases

SourceExternal IDURL
Hugging Face Datasetstimmzak/ReText-Captionshttps://huggingface.co/datasets/timmzak/ReText-Captions Read discussion