MO R-CNN: Multispectral Oriented R-CNN for Object Detection in Remote Sensing Image
This addresses computational efficiency and accuracy challenges in remote sensing object detection, though it appears incremental as it builds on existing R-CNN and large kernel convolution approaches.
The paper tackles the problem of oriented object detection in multi-spectral remote sensing images by proposing MO R-CNN, a lightweight framework that achieves state-of-the-art performance on three datasets (DroneVehicle, VEDAI, and OGSOD).
Oriented object detection for multi-spectral imagery faces significant challenges due to differences both within and between modalities. Although existing methods have improved detection accuracy through complex network architectures, their high computational complexity and memory consumption severely restrict their performance. Motivated by the success of large kernel convolutions in remote sensing, we propose MO R-CNN, a lightweight framework for multi-spectral oriented detection featuring heterogeneous feature extraction network (HFEN), single modality supervision (SMS), and condition-based multimodal label fusion (CMLF). HFEN leverages inter-modal differences to adaptively align, merge, and enhance multi-modal features. SMS constrains multi-scale features and enables the model to learn from multiple modalities. CMLF fuses multimodal labels based on specific rules, providing the model with a more robust and consistent supervisory signal. Experiments on the DroneVehicle, VEDAI and OGSOD datasets prove the superiority of our method. The source code is available at:https://github.com/Iwill-github/MORCNN.