Datasets
Standard Dataset
RoRaTrack
- Citation Author(s):
-
ShreyaGhosh
Purdue University - Submitted by:
- Shreya Ghosh
- Last updated:
- Wed, 02/19/2025 - 13:54
- DOI:
- 10.21227/cr6j-bp98
- Data Format:
- License:
26 Views- Categories:
- Keywords:
Abstract
A significant challenge in racing-related research is the lack of publicly available datasets containing raw images with corresponding annotations for the downstream task. In this paper, we introduce RoRaTrack, a novel dataset that contains annotated multi-camera image data from racing scenarios for track detection. The data is collected on a Dallara AV-21 at a racing circuit in Indiana, in collaboration with the Indy Autonomous Challenge (IAC). RoRaTrack addresses common problems such as blurriness due to high speed, color inversion from the camera, and absence of lane markings on the track. Consequently, we propose RaceGAN, a baseline model based on a Generative Adversarial Network (GAN) that effectively addresses these challenges. The proposed model demonstrates superior performance compared to current state-of-the-art machine learning models in track detection. The dataset and code for this work are available at github.com/RaceGAN.
This dataset contains 1,398 images collected from the Putnam Park Road Course in Indiana. It is designed for track detection tasks, with image data of the racing track and their corresponding label masks, which help in identifying and detecting the track boundaries. This dataset is the one corresponding to the paper "A Racing Dataset and Baseline Model for Track Detection in Autonomous Racing" and provides a valuable resource for training and evaluating machine learning models aimed at autonomous racing and track analysis.
The dataset includes the original training and testing splits, which are provided in the train.txt and test.txt files, respectively. The images offer a variety of track conditions and views, making it an excellent dataset for developing accurate track detection systems.
