1st Place Solution to NeurIPS 2022 Challenge on Visual Domain Adaptation

TL;DR

This paper presents SIA_Adapt, a novel domain adaptation method that combines large-scale pre-training, self-training with pseudo-labels, and model soup techniques, achieving first place in the VisDA 2022 challenge for industrial waste segmentation.

Contribution

Introduces SIA_Adapt, a new domain adaptive model leveraging large-scale pre-training, a unique architecture, self-training, and model soup for improved generalization.

Findings

Achieved 1st place in the VisDA 2022 challenge.

Demonstrated effectiveness of combining pre-training, self-training, and model soup.

Improved domain adaptation performance in industrial waste segmentation.

Abstract

The Visual Domain Adaptation(VisDA) 2022 Challenge calls for an unsupervised domain adaptive model in semantic segmentation tasks for industrial waste sorting. In this paper, we introduce the SIA_Adapt method, which incorporates several methods for domain adaptive models. The core of our method in the transferable representation from large-scale pre-training. In this process, we choose a network architecture that differs from the state-of-the-art for domain adaptation. After that, self-training using pseudo-labels helps to make the initial adaptation model more adaptable to the target domain. Finally, the model soup scheme helped to improve the generalization performance in the target domain. Our method SIA_Adapt achieves 1st place in the VisDA2022 challenge. The code is available on https: //github.com/DaehanKim-Korea/VisDA2022_Winner_Solution.