CamVid¶
Attribute |
Value |
---|---|
pretty_name |
CamVid Dataset |
annotations_creators |
|
language_creators |
|
languages |
|
licenses |
CC BY-NC-ND 4.0 |
multilinguality |
|
size_categories |
100<n<1K |
source_datasets |
|
task_categories |
semantic-segmentation |
task_ids |
|
paperswithcode_id |
camvid |
Dataset Description¶
Homepage: CamVid
Licenses: Attribution-NonCommercial-NoDerivatives 4.0 International
Dataset Summary¶
Cambridge-driving Labeled Video Database: Road/driving scene understanding database.
Download and prepare data¶
Download the data from this github repository and extract it. Replace {PATH_TO_DATA} below with the location of the CamVid folder. Use the following code to load it:
from squirrel_datasets_core.datasets.camvid import CamvidDriver
iter_train = CamvidDriver("{PATH_TO_DATA}").get_iter("train")
iter_val = CamvidDriver("{PATH_TO_DATA}").get_iter("val")
iter_test = CamvidDriver("{PATH_TO_DATA}").get_iter("test")
Dataset Structure¶
Data Instances¶
A sample from the training set is provided below:
{
'image_url': '{PATH_TO_DATA}/train/0016E5_07920.png',
'label_url': '{PATH_TO_DATA}/trainannot/0016E5_07920.png',
'split': 'train'
'image': array(...)
'label': array(...)
}
Dataset Schema¶
img: A numpy array containing the 480x360 RGB image.
label: semantic segmentation map - sky (0), building (1), pole (2), road (3), pavement (4), tree (5), sign/symbol (6), fence (7), car (8), pedestrian (9), bicyclist (10), unlabelled (11)
Data Splits¶
name |
train |
val |
test |
camvid |
367 |
101 |
233 |