Convolutional neural networks have been successfully applied to semantic segmentation problems. However, there are many problems that are inherently not pixel-wise classification problems but are frequently formulated as semantic segmentation. This ill-posed formulation consequently necessitates hand-crafted scenario-specific and computationally expensive post-processing methods to convert the per pixel probability maps to final desired outputs. Generative adversarial networks (GANs) can be used to make the semantic segmentation network output to be more realistic or better structure-preserving, decreasing the dependency on potentially complex post-processing. In this work, we propose EL-GAN: a GAN framework to mitigate the discussed problem using an embedding loss. With EL-GAN, we discriminate based on learned embeddings of both the labels and the prediction at the same time. This results in more stable training due to having better discriminative information, benefiting from seeing both 'fake' and 'real' predictions at the same time that substantially stabilizes the adversarial training process. We use the TuSimple lane marking challenge to demonstrate that using our proposed framework it is viable to overcome the inherent anomalies of posing it as a semantic segmentation problem. Not only is the output considerably more similar to the labels when compared to conventional methods, the subsequent post-processing is simpler and crosses the competitive 96% accuracy threshold.

本文提出了一种名为EL-GAN的基于生成对抗网络的语义分割框架，通过嵌入损失降低了后处理的复杂度和提高了训练稳定性，应用于TuSimple车道线数据集上，效果优于传统方法并达到96%以上的准确率。

EL-GAN: 以嵌入损失为驱动的生成对抗网络用于车道检测