Semi-Supervised Domain Adaptation with Representation Learning for Semantic Segmentation across Time

TL;DR

This paper introduces a semi-supervised domain adaptation technique for semantic segmentation that fine-tunes a network on new datasets using feature map regression, reducing annotation needs while maintaining high performance.

Contribution

It presents a novel domain adaptation approach that leverages feature map regression to adapt models across datasets with similar content but different pixel distributions.

Findings

Achieves performance comparable to transfer learning on PASCAL VOC

Reduces need for pixel-wise annotations in new datasets

Effective for datasets with similar semantic content but different distributions

Abstract

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.