EHSNet: End-to-End Holistic Learning Network for Large-Size Remote Sensing Image Semantic Segmentation
This addresses a critical bottleneck for researchers and practitioners in remote sensing by enabling holistic segmentation of large images, though it appears incremental as it builds on existing segmentation methods with new modules.
The paper tackles the problem of GPU memory exhaustion in large-size remote sensing image semantic segmentation by proposing EHSNet, an end-to-end network that processes holistic images using memory-friendly modules and memory offloading, achieving improvements of +5.65 mIoU on FBP and +4.28 mIoU on Inria Aerial datasets.
This paper presents EHSNet, a new end-to-end segmentation network designed for the holistic learning of large-size remote sensing image semantic segmentation (LRISS). Large-size remote sensing images (LRIs) can lead to GPU memory exhaustion due to their extremely large size, which has been handled in previous works through either global-local fusion or multi-stage refinement, both of which are limited in their ability to fully exploit the abundant information available in LRIs. Unlike them, EHSNet features three memory-friendly modules to utilize the characteristics of LRIs: a long-range dependency module to develop long-range spatial context, an efficient cross-correlation module to build holistic contextual relationships, and a boundary-aware enhancement module to preserve complete object boundaries. Moreover, EHSNet manages to process holistic LRISS with the aid of memory offloading. To the best of our knowledge, EHSNet is the first method capable of performing holistic LRISS. To make matters better, EHSNet outperforms previous state-of-the-art competitors by a significant margin of +5.65 mIoU on FBP and +4.28 mIoU on Inria Aerial, demonstrating its effectiveness. We hope that EHSNet will provide a new perspective for LRISS. The code and models will be made publicly available.