naver-ai/eccv-caption

Extended COCO Validation (ECCV) Caption dataset (ECCV 2022)

38
/ 100
Emerging

When developing or evaluating AI models that understand both images and text, a common challenge is ensuring the model correctly associates an image with its relevant captions, and vice-versa. This project provides an extended dataset and a toolkit to measure how accurately your image-text model performs these associations. It takes your model's ranked lists of captions for an image, or images for a caption, and outputs a suite of performance metrics. This is for researchers and engineers building and benchmarking multimodal AI models.

No commits in the last 6 months. Available on PyPI.

Use this if you are evaluating the performance of your image-text matching AI model and need more accurate, human- and machine-verified ground truth data beyond the original COCO Caption dataset, along with standardized metrics.

Not ideal if you are a casual user looking for an out-of-the-box image captioning or image search solution; this is a toolkit for model evaluation, not a deployed application.

image-text-matching multimodal-AI model-evaluation computer-vision natural-language-processing
Stale 6m No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 25 / 25
Community 5 / 25

How are scores calculated?

Stars

56

Forks

2

Language

Python

License

Last pushed

Mar 01, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/naver-ai/eccv-caption"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.