IDD-3D: Indian Driving Dataset for 3D Unstructured Road Scenes

TL;DR

The paper introduces IDD-3D, a large-scale, multi-modal dataset capturing complex unstructured road scenes in India, aimed at improving autonomous driving models' adaptability to diverse and challenging real-world traffic conditions.

Contribution

The creation of a comprehensive, geographically diverse 3D driving dataset from India with multi-modal data, addressing limitations of existing datasets biased towards developed cities.

Findings

IDD-3D contains 12,000 annotated LiDAR frames.

Benchmark results on 3D detection and tracking demonstrate the dataset's complexity.

Statistical analysis shows IDD-3D's diversity exceeds existing datasets.

Abstract

Autonomous driving and assistance systems rely on annotated data from traffic and road scenarios to model and learn the various object relations in complex real-world scenarios. Preparation and training of deploy-able deep learning architectures require the models to be suited to different traffic scenarios and adapt to different situations. Currently, existing datasets, while large-scale, lack such diversities and are geographically biased towards mainly developed cities. An unstructured and complex driving layout found in several developing countries such as India poses a challenge to these models due to the sheer degree of variations in the object types, densities, and locations. To facilitate better research toward accommodating such scenarios, we build a new dataset, IDD-3D, which consists of multi-modal data from multiple cameras and LiDAR sensors with 12k annotated driving LiDAR frames across various traffic scenarios. We discuss the need for this dataset through statistical comparisons with existing datasets and highlight benchmarks on standard 3D object detection and tracking tasks in complex layouts. Code and data available at https://github.com/shubham1810/idd3d_kit.git