DepthFormer: Multimodal Positional Encodings and Cross-Input Attention for Transformer-based Segmentation Networks.

Francesco Barbato, Giulia Rizzoli, Pietro Zanuttigh

Browse the full ICASSP paper archive.