Synthetic Convolutional Features for Improved Semantic Segmentation
This work addresses the problem of enhancing semantic segmentation accuracy for computer vision applications, representing an incremental advancement in leveraging synthetic data.
The paper tackles the challenge of using synthetic data to improve semantic segmentation by proposing a method to generate intermediate convolutional features from label masks, which are then incorporated into training. Experimental results on Cityscapes and ADE20K datasets show that this approach improves segmentation performance.
Recently, learning-based image synthesis has enabled to generate high-resolution images, either applying popular adversarial training or a powerful perceptual loss. However, it remains challenging to successfully leverage synthetic data for improving semantic segmentation with additional synthetic images. Therefore, we suggest to generate intermediate convolutional features and propose the first synthesis approach that is catered to such intermediate convolutional features. This allows us to generate new features from label masks and include them successfully into the training procedure in order to improve the performance of semantic segmentation. Experimental results and analysis on two challenging datasets Cityscapes and ADE20K show that our generated feature improves performance on segmentation tasks.