Semantic scene segmentation has become a key application in computer vision and is an essential part of intelligent transportation systems for complete scene understanding of the surrounding environment. While several methods based on deep fully Convolutional Neural Network (CNN) have been emerging, there are two main challenges: (i) They mainly focus on improvement of the accuracy than efficiency. (ii) They assume structured driving environment like in USA and Europe. While most of the current works focus on the well structured driving environment, we focus our research on India Driving Dataset (IDD) which contains data from unstructured traffic scenario. In this paper, we propose modifications in the DeepLabV3+ framework by using lower atrous rates in Atrous Spatial Pyramid Pooling (ASPP) module for dense traffic prediction. We propose to use dilated Xception network as the backbone for feature extraction. A lightweight segmentation framework is also presented by exploring the effectiveness of MobileNetV2 architecture, which achieves competitively high accuracy and is much smaller than other state-of-art architectures. The performance is evaluated in terms of mean Intersection over Union (mIoU) on 26 fine grained classes of IDD. Our proposed model with 24 M parameters achieves 68.41 mIoU on test set and efficient mobile model achieves mIoU of 61.6 by reducing the parameters to 2.2 M only.

语义场景分割已成为计算机视觉中的关键应用程序，并且是智能交通运输系统中必不可少的部分，可以对周围环境进行完整的场景理解。尽管已经出现了几种基于深度完全卷积神经网络（CNN）的方法，但存在两个主要挑战：（i）它们主要集中在提高准确性而不是效率上。（ii）他们假设结构化的驾驶环境如美国和欧洲。尽管当前的大多数工作都集中在结构化的驾驶环境上，但我们的研究重点是印度驾驶数据集（IDD），其中包含来自非结构化交通场景的数据。在本文中，我们建议在DeepLabV3 +框架中进行修改，方法是使用Atrous空间金字塔池（ASPP）模块中较低的atrous速率进行密集流量预测。我们建议使用膨胀的Xception网络作为特征提取的主干。通过探索MobileNetV2架构的有效性，还提出了一种轻量级的细分框架，该架构具有竞争性的高精度，并且比其他最新架构更小。该性能是根据IDD的26个细粒度类别上的平均联盟交集（mIoU）进行评估的。我们提出的具有24 M参数的模型在测试集上达到68.41 mIoU，而有效的移动模型通过将参数减少到仅2.2 M来达到61.6 mIoU。该性能是根据IDD的26个细粒度类别上的平均联盟交集（mIoU）进行评估的。我们提出的具有24 M参数的模型在测试集上达到68.41 mIoU，而有效的移动模型通过将参数减少到仅2.2 M来达到61.6 mIoU。该性能是根据IDD的26个细粒度类别上的平均联盟交集（mIoU）进行评估的。我们提出的具有24 M参数的模型在测试集上达到68.41 mIoU，而有效的移动模型通过将参数减少到仅2.2 M来达到61.6 mIoU。