Self Semi Supervised Neural Architecture Search for Semantic Segmentation
This work addresses the need for automated and efficient neural network design in semantic segmentation, offering a domain-specific improvement that is incremental in nature.
The paper tackles the problem of designing efficient neural networks for semantic segmentation by proposing a neural architecture search strategy that combines self-supervised and semi-supervised learning, resulting in a discovered model that is four times more efficient in floating operations than a state-of-the-art hand-crafted model.
In this paper, we propose a Neural Architecture Search strategy based on self supervision and semi-supervised learning for the task of semantic segmentation. Our approach builds an optimized neural network (NN) model for this task by jointly solving a jigsaw pretext task discovered with self-supervised learning over unlabeled training data, and, exploiting the structure of the unlabeled data with semi-supervised learning. The search of the architecture of the NN model is performed by dynamic routing using a gradient descent algorithm. Experiments on the Cityscapes and PASCAL VOC 2012 datasets demonstrate that the discovered neural network is more efficient than a state-of-the-art hand-crafted NN model with four times less floating operations.