README.txt
==========
This is a JSON format of the Human dataset from the paper "'Where am I?' Scene Retrieval with Language" from ECCV 2024. The "description" is human generated, the "scanId" is the corresponding scan id from the 3DSSG dataset [1].

[Our project page: https://whereami-langloc.github.io/]

    - list[dict]:                               List of scan descriptions

        - description (str):                    Text description of a scene, written by a human annotator, describing the objects and object relationships in the scene.

        - scanId (str):                         String representing the 3DSSG scan ID that corresponds to this description.


[1] Wald, J., Dhamo, H., Navab, N., Tombari, F.: Learning 3d semantic scene graphs from 3d indoor reconstructions (CVPR 2020) 

[3DSSG project page: https://3dssg.github.io/]


If you consider this dataset useful, please consider citing our paper:
@article{chen2024whereami,
  author    = {Chen, Jiaqi and Barath, Daniel and Armeni, Iro and Pollefeys, Marc and Blum, Hermann},
  title     = {"Where am I?" Scene Retrieval with Language},
  journal   = {ECCV},
  year      = {2024},
}