Learning Modality Geometry for Multispectral Object Detection: Adaptive Spacing and 3-D Rotary Position Embedding (opens in new tab)

Positional encoding is crucial for the Transformer to effectively process multimodal feature information in multispectral object detection. However, existing studies often directly apply single-modal encoding designs to multimodal tasks, resulting in positional encoding focusing on spatial information while neglecting modality-specific information, which limits the performance of the Transformer. To explore the potential of positional encoding in jointly processing spatial and modality inform...