IndiaScene365: a transfer learning dataset for Indian scene understanding in diverse weather condition
detail [15]. This limits the ability to train models 111 that can generalize well to such conditions. Foggy 112 Zurich [3] and Foggy Cityscapes [4] introduced 113 synthetic fog images by applying masks to original 114 images. Synthetic rain datasets [14] include 115 Rain1400, RainyCityscapes Rain100H, and 116 Rain12. Datasets like, BDD100K, ACDC [11], and 117 IDD [1], contain authentic images collected from 118 various adverse conditions [15], including fog, 119 rain, snow, and nighttime scenarios [5] [6]. The label set used in dataset preparation is identical to that in IDD [1] The hierarchy in the category of class labels adds a higher degree of complexity [5] to our dataset compared to existing datasets like Cityscapes [7], and even when compared to adverse weather datasets such as Foggy Cityscapes [8]
Authors
- Sachin D. Shelke
- Sandhya Arora
- Deepa Mane
Institutions
- Indian Institute of Information Technology, Pune (IN)
- Smt. Kashibai Navale Medical College and General hospital (IN)
- Savitribai Phule Pune University (IN)
Publication Details
- Journal
- Frontiers in Artificial Intelligence
- Published
- 2026-01-09
- DOI
- https://doi.org/10.3389/frai.2025.1669512
- Primary Topic
- Image Enhancement Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00