labelme Image Dataset
Versions
editedclass
v17
Oct 9, 2023
2023-10-09 7:47am
v16
Oct 9, 2023
Original
v15
Sep 15, 2023
2023-08-24 5:35am
v14
Aug 24, 2023
2023-08-20 11:31am
v13
Aug 20, 2023
2023-08-20 11:06am
v12
Aug 20, 2023
2023-08-19 11:17pm
v11
Aug 19, 2023
2023-08-19 4:02pm
v10
Aug 19, 2023
originalclass
v9
Jul 30, 2023
2023-07-29 1:51pm
v8
Jul 29, 2023
2023-07-29 1:45pm
v7
Jul 29, 2023
2023-07-11 12:01pm
v6
Jul 11, 2023
2023-06-26 10:53pm
v5
Jun 26, 2023
2023-06-26 10:41pm
v4
Jun 26, 2023
2023-06-26 10:14pm
v3
Jun 26, 2023
2023-06-25 9:01pm
v2
Jun 25, 2023
v17
editedclass
Generated on Oct 9, 2023
797 Total Images
View All Images
Dataset Split
Train Set 75%
597Images
Valid Set 23%
180Images
Test Set 3%
20Images
Preprocessing
Resize: Stretch to 1000x667
Modify Classes: 2 remapped, 7 dropped
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
Saturation: Between -25% and +25%
Brightness: Between -25% and +25%
Similar Projects
urban-elements
399 images
unlabeled-paved-area-dirt-grass
399 images
test
400 images
cars-persons-trucks-vehicles-
400 images
humans
347 images