Hybrid transformer-CNN networks using superpixel segmentation for remote sensing building change detection
Sign Up to like & getrecommendations! Published in 2023 at "International Journal of Remote Sensing"
DOI: 10.1080/01431161.2023.2208711
Abstract: ABSTRACT Convolution in convolutional neural network(CNN) essentially uses a filter (kernel) with shared parameters to achieve feature extraction by computing the weighted sum of the centre pixel and adjacent pixels. The transformer divides the input… read more here.
Keywords: hybrid transformer; remote sensing; transformer cnn; cnn ... See more keywords
Transformer and CNN Hybrid Deep Neural Network for Semantic Segmentation of Very-high-resolution Remote Sensing Imagery
Sign Up to like & getrecommendations! Published in 2022 at "IEEE Transactions on Geoscience and Remote Sensing"
DOI: 10.1109/tgrs.2022.3144894
Abstract: This paper presents a transformer and CNN hybrid deep neural network for semantic segmentation of very-high-resolution remote sensing imagery. The model follows an encoder-decoder structure. The encoder module uses a new universal backbone swin transformer… read more here.
Keywords: network; cnn hybrid; remote sensing; transformer cnn ... See more keywords
Transformer-CNN: Swiss knife for QSAR modeling and interpretation
Sign Up to like & getrecommendations! Published in 2020 at "Journal of Cheminformatics"
DOI: 10.1186/s13321-020-00423-w
Abstract: We present SMILES-embeddings derived from the internal encoder state of a Transformer [ 1 ] model trained to canonize SMILES as a Seq2Seq problem. Using a CharNN [ 2 ] architecture upon the embeddings results… read more here.
Keywords: cnn swiss; qsar; swiss knife; transformer cnn ... See more keywords