You can deploy this model and start earning money today!

SInadumi/
mmrr

Name: mmrr
Availability: InStock

Views

No views yet

J-CRe3 Checkpoints and Image Subset

This directory provides:

Pretrained model checkpoints for phrase grounding and multimodal reference resolution.
A subset of the J-CRe3 dataset containing only images and metadata (e.g., info.json).

📁 Directory Structure of J-CRe3 subset

bash

datasets/
	└─ J-CRe3/
		└─ recording/
			└─ 20220302-56130207-1/
				├── info.json
				└── images/
					├── 001.png
					├── 002.png
					└── ...
			└─ 20220302-56130218-2/
			└─ ...

Please place the recording/ folder under the annotation directory of the J-CRe3 visual/textual annotations, which can be downloaded from here.

Acknowledgement

License

The license for this checkpoints and data are subject to CC BY-SA 4.0.

SInadumi/mmrr

J-CRe3 Checkpoints and Image Subset

📁 Directory Structure of J-CRe3 subset

Acknowledgement

License

SInadumi/
mmrr