BCG-Former: Toward Pareto-Efficient Hyperspectral Image Classification via Band-Contextual Gating
Computer Vision and Pattern Recognition
Summary
The authors developed BCG-Former, a lightweight model that combines convolutional neural networks and transformers to classify hyperspectral images efficiently. Their model balances accuracy with low memory use and very fast processing speeds, making it suitable for devices like drones or small satellites with limited computing power. They introduced three key features that help the model understand complex spectral data better and tested it on several well-known datasets, showing strong performance with fewer parameters and quicker inference than other methods. Their analysis showed all parts of their design help, especially the Band-Contextual Gating mechanism. Overall, the authors present BCG-Former as a practical solution for accurate and efficient hyperspectral imaging analysis in real-time settings.
Authors
Gaurav Sharma, Eungjoo Lee
Abstract
Hyperspectral image (HSI) classification systems are increasingly deployed on platforms with strict computational budgets, such as UAVs and small spaceborne sensors. In these settings, accuracy alone is not enough; the model must also run within tight latency and memory constraints. Most recent HSI classifiers, however, focus on accuracy and pay relatively little attention to these constraints. We propose BCG-Former, a lightweight CNN-Transformer hybrid that targets this trade-off. The model introduces three innovations: (1) Band-Contextual Gating (BCG) for adaptive spectral recalibration using local inter-band context and learnable temperature sharpening, (2) a spectral summary token that bridges spectral and spatial features, and (3) single-pass Band-RoPE combined with linear attention for efficient joint representation learning. Evaluated on classical airborne (Pavia University, Salinas, Indian Pines, Houston 2013/2018) and UAV-borne benchmark datasets (WHU-Hi-LongKou, HongHu, and HanChuan), BCG-Former achieves over-all accuracy ranging from 91.51% on Houston 2018 to 99.49% on Houston 2013, while maintaining sub-millisecond inference latency (0.91-0.95ms) and using only 0.10-0.23M parameters. Across all eight benchmarks, BCG-Former consistently resides on or near the Pareto frontier of accuracy versus latency, outperforming or matching recent CNN-, Transformer-, and Mamba-based methods at a fraction of their computational cost. Ablation studies confirm that all three components are complementary, with BCG providing the largest individual contribution. These results establish BCG-Former as a strong accuracy-efficiency Pareto candidate for real-time and large-scale remote sensing applications.