Multi-Scale Semantic Segmentation with Modified MBConv Blocks
This work addresses the need for efficient and effective semantic segmentation in computer vision, though it is incremental as it adapts existing blocks to a new task.
The paper tackled the problem of adapting MBConv blocks for semantic segmentation by modifying them to extract detailed spatial information across scales, achieving mean IoU scores of 84.5% and 84.0% on Cityscapes test and validation datasets.
Recently, MBConv blocks, initially designed for efficiency in resource-limited settings and later adapted for cutting-edge image classification performances, have demonstrated significant potential in image classification tasks. Despite their success, their application in semantic segmentation has remained relatively unexplored. This paper introduces a novel adaptation of MBConv blocks specifically tailored for semantic segmentation. Our modification stems from the insight that semantic segmentation requires the extraction of more detailed spatial information than image classification. We argue that to effectively perform multi-scale semantic segmentation, each branch of a U-Net architecture, regardless of its resolution, should possess equivalent segmentation capabilities. By implementing these changes, our approach achieves impressive mean Intersection over Union (IoU) scores of 84.5% and 84.0% on the Cityscapes test and validation datasets, respectively, demonstrating the efficacy of our proposed modifications in enhancing semantic segmentation performance.