Rank-Aware Quantile Activation for Motion-Robust Crop Segmentation in UAV Imagery
For UAV-based crop segmentation, this work addresses the problem of motion blur degrading performance on rare agronomic classes, offering a complementary robustness source to blur-domain training.
Motion blur from UAVs degrades semantic segmentation on rare texture-dependent classes. The proposed Dual Quantile Activation (QAct) achieves consistent mIoU gains over ReLU across zero-shot and blur-supervised regimes, with strongest gains on rare classes; at moderate blur, zero-shot QAct outperforms distillation-trained ReLU.
Motion blur from high-speed UAV acquisition de-grades semantic segmentation on rare texture-dependent classes with high agronomic value. Standard CNNs rely on high-frequency magnitude features that blur destroys, causing statistical erasure of minority signals. We propose Dual Quantile Activation (QAct), a rank-aware block replacing magnitude gating with instance-level rank normalization. Evaluated onAgriculture-Vision 2021 across zero-shot and blur-supervised regimes at multiple severities, QAct is the dominant architectural factor: it delivers consistent mIoU gains over ReLU across both regimes and all severities, with strongest gains on rare structural and texture-dependent classes. Some dominant classes (water,planter skip) show mixed per-class performance under distillation. At moderate blur, zero-shot QAct outperforms distillation-trained ReLU; across all severities, Distill-QAct achieves best performance, confirming rank aware activation and blur-domain training are complementary robustness sources.