ShadowCast improves remote sensing segmentation by adding synthetic shadows
ShadowCast: On-the-Fly Context-Aware Shadow Augmentation for Remote Sensing Image Segmentation
Computer Vision and Pattern Recognition
Summary
Shadows in satellite images make it hard for computers to tell what objects are there. Instead of trying to remove shadows, the authors created ShadowCast, which adds fake shadows to training images to help the computer learn better. ShadowCast figures out where shadows should fall based on the objects and brightness around them without needing extra data or complicated models. Tests show that adding these fake shadows helps improve object recognition in satellite pictures across several datasets.
What this means in practice
- •For remote sensing engineers: Improve building and tree segmentation models by incorporating synthesized shadows to boost accuracy without extra shadow data or models.
- •For geospatial analytics teams: Enhance land use mapping tools by training segmentation systems that handle shadows better with this simple augmentation method.
Authors
Yechan Kim, Dongho Yoon, Jongmin Joo, JongHyun Park, Moongu Jeon
Abstract
Shadows cast by elevated objects alter the spectral appearance and obscure the boundaries of objects in remote sensing images, thereby degrading semantic segmentation performance. Most existing approaches handle shadows as degradations to be detected and removed, requiring additional costs from dedicated networks and shadow-specific annotations. We propose ShadowCast, which takes the opposite approach: it synthesizes shadows and injects them during training for remote sensing segmentation. It extracts masks of elevated objects, such as buildings and trees, directly from the segmentation labels and estimates an image-specific cast-shadow direction cue by probing the brightness around them. It then projects synthetic shadows along the inferred cast-shadow direction, requiring no learned network and no acquisition metadata. Experiments on three widely used remote sensing benchmarks, Vaihingen, Potsdam, and LoveDA, show that ShadowCast consistently outperforms the baseline across diverse state-of-the-art segmentation architectures. Moreover, it brings additional performance gains when combined with standard augmentation and regularization techniques, indicating that stochastic shadow injection serves as a simple and complementary augmentation strategy for remote sensing segmentation.