Memory‐augmented neural networks based dynamic complex image segmentation in digital twins for self‐driving vehicle

作者：

Highlights：

• An image segmentation model is built on Memory-augmented Neural Networks (MANNs), in an effort to identify the ever-boosting image information with more details.

• Results demonstrate that the MANNs-based image segmentation model is more accurate and

• consumes less training time than other classic models.

摘要

•An image segmentation model is built on Memory-augmented Neural Networks (MANNs), in an effort to identify the ever-boosting image information with more details.•Results demonstrate that the MANNs-based image segmentation model is more accurate and•consumes less training time than other classic models.

论文关键词：Deep learning,Image segmentation,Memory-augmented neural networks,LSTM,Self-driving, Digital twins

论文评审过程：Received 10 October 2021, Revised 13 June 2022, Accepted 2 August 2022, Available online 6 August 2022, Version of Record 10 August 2022.

论文官网地址：https://doi.org/10.1016/j.patcog.2022.108956