Post

Good at captioning, bad at counting Benchmarking GPT-4V on Earth observation data

Posted
By 1 min read
  • 论文名称: Good at captioning, bad at counting: Benchmarking GPT-4V on Earth observation data
  • 模型架构: Other
  • Link: https://arxiv.org/abs/2401.17600
  • Code/Project: https://vleo.danielz.ch/
  • Short Summary: 提出了一个benchmark，用于评估VLMs在地球观测数据上的表现，特别是在场景理解、定位和计数以及变化检测任务上
  • Published in: Arxiv 2023
  • 备注: 评估
Arxiv
paper Other
This post is licensed under CC BY 4.0 by the author.