Indian Road Driver’s View Dataset

Despite the existence of road image datasets, these datasets predominantly focus on European roads with less variability in traffic and road conditions. To address this limitation, we have developed an image dataset tailored to Indian road conditions, capturing the extensive variations in traffic and environment. Our dataset, the Indian Road Driver's View Dataset, was created under clear daylight conditions across diverse locations including Shillong, Mumbai, Kolkata, Kanpur, Jaipur, Hyderabad, Guwahati, Delhi, and Chennai.  This dataset provides a comprehensive collection of road images from various urban, rural, and highway settings in India, enhancing the representation of different traffic conditions. The dataset was captured using smartphone cameras mounted on car bonnets, utilizing auto settings to ensure well-exposed images. Environmental parameters such as temperature, humidity, and luminosity were recorded during image capture. We benchmarked our physics-inspired darkening models against existing European-focused datasets and our newly collected Indian dataset, demonstrating the significance of including varied traffic conditions. This contribution is crucial for developing more robust and generalizable models for road scene analysis in diverse traffic environments.