Semi-Supervised Domain Adaptation with Representation Learning for Semantic Segmentation across Time
This addresses the data collection bottleneck for researchers and practitioners in computer vision, but it is incremental as it builds on existing domain adaptation and transfer learning approaches.
The paper tackles the problem of expensive data annotation for semantic segmentation by proposing a semi-supervised domain adaptation method that finetunes a pre-trained network on a new dataset without requiring annotations, achieving performances similar to classic transfer learning on the PASCAL VOC dataset with synthetic transformations.
Deep learning generates state-of-the-art semantic segmentation provided that a large number of images together with pixel-wise annotations are available. To alleviate the expensive data collection process, we propose a semi-supervised domain adaptation method for the specific case of images with similar semantic content but different pixel distributions. A network trained with supervision on a past dataset is finetuned on the new dataset to conserve its features maps. The domain adaptation becomes a simple regression between feature maps and does not require annotations on the new dataset. This method reaches performances similar to classic transfer learning on the PASCAL VOC dataset with synthetic transformations.