Efficient Semantic Segmentation using Gradual Grouping
This work addresses efficiency challenges for real-time semantic segmentation in applications like autonomous driving, though it is incremental as it builds on existing techniques.
The paper tackled the problem of high memory and runtime in deep CNNs for semantic segmentation by applying efficient techniques like grouped and depth-wise separable convolutions to ERFNet, achieving a 5X improvement in FLOPs with only a 4% accuracy degradation on the Cityscapes dataset.
Deep CNNs for semantic segmentation have high memory and run time requirements. Various approaches have been proposed to make CNNs efficient like grouped, shuffled, depth-wise separable convolutions. We study the effectiveness of these techniques on a real-time semantic segmentation architecture like ERFNet for improving run time by over 5X. We apply these techniques to CNN layers partially or fully and evaluate the testing accuracies on Cityscapes dataset. We obtain accuracy vs parameters/FLOPs trade offs, giving accuracy scores for models that can run under specified runtime budgets. We further propose a novel training procedure which starts out with a dense convolution but gradually evolves towards a grouped convolution. We show that our proposed training method and efficient architecture design can improve accuracies by over 8% with depth wise separable convolutions applied on the encoder of ERFNet and attaching a light weight decoder. This results in a model which has a 5X improvement in FLOPs while only suffering a 4% degradation in accuracy with respect to ERFNet.