Local-to-Global Information Communication for Real-Time Semantic Segmentation Network Search
This work addresses the need for efficient and accurate real-time semantic segmentation models, particularly for applications like autonomous driving, by improving neural architecture search methods, though it is incremental in nature.
The paper tackled the problem of automatically designing neural network architectures for real-time semantic segmentation by introducing a new search space with cell-independent cells and modules for local-to-global information communication, achieving a new state-of-the-art trade-off with 74.0% mIoU at 115.2 FPS on Cityscapes.
Neural Architecture Search (NAS) has shown great potentials in automatically designing neural network architectures for real-time semantic segmentation. Unlike previous works that utilize a simplified search space with cell-sharing way, we introduce a new search space where a lightweight model can be more effectively searched by replacing the cell-sharing manner with cell-independent one. Based on this, the communication of local to global information is achieved through two well-designed modules. For local information exchange, a graph convolutional network (GCN) guided module is seamlessly integrated as a communication deliver between cells. For global information aggregation, we propose a novel dense-connected fusion module (cell) which aggregates long-range multi-level features in the network automatically. In addition, a latency-oriented constraint is endowed into the search process to balance the accuracy and latency. We name the proposed framework as Local-to-Global Information Communication Network Search (LGCNet). Extensive experiments on Cityscapes and CamVid datasets demonstrate that LGCNet achieves the new state-of-the-art trade-off between accuracy and speed. In particular, on Cityscapes dataset, LGCNet achieves the new best performance of 74.0\% mIoU with the speed of 115.2 FPS on Titan Xp.