Few-Shot Rotation-Invariant Aerial Image Semantic Segmentation
This solves the problem of accurately segmenting rotated objects in aerial images with limited data, which is incremental as it builds on existing few-shot methods by adding rotation invariance.
The paper tackles the problem of few-shot aerial image segmentation by addressing failures due to varying object orientations, proposing FRINet to match features rotation-adaptively, and achieves state-of-the-art performance on benchmarks.
Few-shot aerial image segmentation is a challenging task that involves precisely parsing objects in query aerial images with limited annotated support. Conventional matching methods without consideration of varying object orientations can fail to activate same-category objects with different orientations. Moreover, conventional algorithms can lead to false recognition of lower-scored rotated semantic objects. In response to these challenges, the authors propose a novel few-shot rotation-invariant aerial semantic segmentation network (FRINet). FRINet matches each query feature rotation-adaptively with orientation-varying yet category-consistent support information. The segmentation predictions from different orientations are supervised by the same label, and the backbones are pre-trained in the base category to boost segmentation performance. Experimental results demonstrate that FRINet achieves state-of-the-art performance in few-shot aerial semantic segmentation benchmark.