- Descripción :
El conjunto de datos de NYU-Depth V2 se compone de secuencias de video de una variedad de escenas de interiores grabadas por las cámaras RGB y de profundidad de Microsoft Kinect.
Documentación adicional : Explore en Papers With Code
Página de inicio: https://cs.nyu.edu/~silberman/datasets/nyu_ depth_v2.html
Código fuente :
tfds.datasets.nyu_depth_v2.BuilderVersiones :
-
0.0.1(predeterminado): Sin notas de la versión.
-
Tamaño de descarga :
31.92 GiBTamaño del conjunto de datos :
74.03 GiBAlmacenamiento automático en caché ( documentación ): No
Divisiones :
| Separar | Ejemplos |
|---|---|
'train' | 47,584 |
'validation' | 654 |
- Estructura de características :
FeaturesDict({
'depth': Tensor(shape=(480, 640), dtype=float16),
'image': Image(shape=(480, 640, 3), dtype=uint8),
})
- Documentación de características :
| Rasgo | Clase | Forma | Tipo D | Descripción |
|---|---|---|---|---|
| CaracterísticasDict | ||||
| profundidad | Tensor | (480, 640) | flotar16 | |
| imagen | Imagen | (480, 640, 3) | uint8 |
Teclas supervisadas (ver documento
as_supervised):('image', 'depth')Figura ( tfds.show_examples ):

- Ejemplos ( tfds.as_dataframe ):
- Cita :
@inproceedings{Silberman:ECCV12,
author = {Nathan Silberman, Derek Hoiem, Pushmeet Kohli and Rob Fergus},
title = {Indoor Segmentation and Support Inference from RGBD Images},
booktitle = {ECCV},
year = {2012}
}
@inproceedings{icra_2019_fastdepth,
author = {Wofk, Diana and Ma, Fangchang and Yang, Tien-Ju and Karaman, Sertac and Sze, Vivienne},
title = {FastDepth: Fast Monocular Depth Estimation on Embedded Systems},
booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
year = {2019}
}