Datasets
Standard Dataset
LaneVisionIITR
- Citation Author(s):
-
SauravKumar
Department of Electrical Engg., IIT RoorkeeSumathiParasuramanDepartment of Electrical Engg., IIT Roorkee - Submitted by:
- Parasuraman Sumathi
- Last updated:
- Mon, 08/07/2023 - 08:05
- DOI:
- 10.21227/aa0c-ha58
- Data Format:
- License:
429 Views- Categories:
- Keywords:
Abstract
"LaneVisionIITR: A Comprehensive High-Resolution Dataset for Lane Detection recorded at IIT Roorkee ", which is a newly built high-resolution dataset for developing Lane detection dataset for advanced driver assistance systems.
This folder consists of three files for each image:
1. The image captured in .jpg format.
2. Annotations (.json) having left and center line coordinates represented as “L” and “C” respectively.
3. The mask image with dots (.jpg) representing two points of the left and two points of center coordinates. Note: The right lane is not considered for annotations since the dataset is prepared for ego-lane line analysis.
Total images: Dataset contains 1000 images in .jpg format.
Annotation: All the images are precisely annotated.
Scenes: Day and night conditions.
Tasks: It can be used for training as well as testing the Lane detection algorithm for advanced driver assistance systems.
High resolution: Image resolution is 1920 X 1080 pixels.
Diversity: A large variation in appearance, scale, and illumination.
The training imageset is too large (442 MB), so it has been zipped. One should download and open the IITR LaneVision.zip using your favourite zip file extractor.
Documentation
| Attachment | Size |
|---|---|
 readme.pdf | 40 KB |
Comments
I am interested in doing a project on this topic