Importance-Aware Semantic Segmentation with Efficient Pyramidal Context Network for Navigational Assistant Systems
This work addresses the need for customized semantic segmentation in navigational assistant systems, such as for autonomous vehicles and visually impaired pedestrians, by incorporating hierarchical importance of objects, though it is incremental in nature.
The paper tackled the problem of semantic segmentation for navigational assistant systems by introducing an importance-aware loss function and extending ERF-PSPNet to BiERF-PSPNet, resulting in high-quality segmentation maps suitable for autonomous vehicles and other applications, as demonstrated on CamVid and Cityscapes datasets.
Semantic Segmentation (SS) is a task to assign semantic label to each pixel of the images, which is of immense significance for autonomous vehicles, robotics and assisted navigation of vulnerable road users. It is obvious that in different application scenarios, different objects possess hierarchical importance and safety-relevance, but conventional loss functions like cross entropy have not taken the different levels of importance of diverse traffic elements into consideration. To address this dilemma, we leverage and re-design an importance-aware loss function, throwing insightful hints on how importance of semantics are assigned for real-world applications. To customize semantic segmentation networks for different navigational tasks, we extend ERF-PSPNet, a real-time segmenter designed for wearable device aiding visually impaired pedestrians, and propose BiERF-PSPNet, which can yield high-quality segmentation maps with finer spatial details exceptionally suitable for autonomous vehicles. A comprehensive variety of experiments with these efficient pyramidal context networks on CamVid and Cityscapes datasets demonstrates the effectiveness of our proposal to support diverse navigational assistant systems.