Multi-Source Soft Pseudo-Label Learning with Domain Similarity-based Weighting for Semantic Segmentation
This work addresses domain adaptation in semantic segmentation for computer vision applications, but it is incremental as it builds on prior multi-source methods.
The paper tackles domain adaptation for semantic segmentation by proposing a soft pseudo-label generation method that weights predictions from multiple source models based on domain similarity to the target dataset, and a training method using entropy to reduce misclassification influence, achieving comparative or better performance than existing methods.
This paper describes a method of domain adaptive training for semantic segmentation using multiple source datasets that are not necessarily relevant to the target dataset. We propose a soft pseudo-label generation method by integrating predicted object probabilities from multiple source models. The prediction of each source model is weighted based on the estimated domain similarity between the source and the target datasets to emphasize contribution of a model trained on a source that is more similar to the target and generate reasonable pseudo-labels. We also propose a training method using the soft pseudo-labels considering their entropy to fully exploit information from the source datasets while suppressing the influence of possibly misclassified pixels. The experiments show comparative or better performance than our previous work and another existing multi-source domain adaptation method, and applicability to a variety of target environments.