Underwater segmentation improved by geometry guided feature enhancement
GeoCueFormer: Geometry-Guided Wavelet Representation and Prediction-Cued Dual-Stage Decoder for Underwater Semantic Segmentation
Computer Vision and Pattern Recognition
Summary
Underwater images are hard to analyze because water changes how things look, making colors faded and edges blurry. The authors built a system called GeoCueFormer that uses information about shapes and depth to better separate important structures from underwater visual noise. This system improves the clarity of details and refines uncertain parts of the image, leading to better recognition of objects underwater. They tested GeoCueFormer on standard underwater datasets and found it outperformed other methods while keeping the model efficient.
What this means in practice
- •For marine monitoring teams: Enhance underwater image analysis to better identify marine species and habitats despite poor visibility conditions.
- •For underwater robotics engineers: Improve autonomous underwater vehicle perception by refining object segmentation in visually degraded underwater environments.
Authors
Xian Wu, Xinjin Li, Yiliu Xu, Yining Liu, Yong Jiang
Abstract
Underwater semantic segmentation is essential for marine ecosystem monitoring, yet remains challenging due to severe visual degradation. Light absorption and scattering often lead to color shifts, low contrast, and blurred boundaries, making shallow detail features unreliable. Existing underwater segmentation methods improve RGB feature aggregation or boundary prediction, but still lack an explicit mechanism to distinguish structure-related details from degradation-induced responses. To address this limitation, we propose GeoCueFormer, a lightweight framework that combines geometry-constrained frequency enhancement with prediction-cued refinement. GeoCueFormer performs stage-specific wavelet enhancement on hierarchical encoder features to complement shallow boundary details while preserving deep structural semantics. A depth-derived spatial gate constrains shallow frequency enhancement toward geometry-consistent regions, and a prediction-cued dual-stage decoder further refines ambiguous high-resolution features. GeoCueFormer obtains 82.23% and 73.04% mIoU on SUIM and DUT, respectively. Under comparable model complexity and standard benchmark settings on SUIM and DUT, it achieves SOTA performance while maintaining a favorable accuracy-complexity trade-off. These results show that distinguishing structural details from degradation-induced interference is more effective for underwater segmentation.