# Alleviating Semantic-level Shift: A Semi-supervised Domain Adaptation Method for Semantic Segmentation

arXiv: https://arxiv.org/abs/2004.00794 · HTML: https://mlsanity.com/paper/2004.00794/

## Key facts

- Cited by 35 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Apr 2020.

## Built on by (most cited first, 30 of 35)

- [Semi-Supervised Semantic Segmentation with Pixel-Level Contrastive Learning from a Class-wise Memory Bank](https://mlsanity.com/paper/2104.13415/) (2021)
- [Cross-modal Learning for Domain Adaptation in 3D Semantic Segmentation](https://mlsanity.com/paper/2101.07253/) (2021)
- [Adaptive Consistency Regularization for Semi-Supervised Transfer Learning](https://mlsanity.com/paper/2103.02193/) (2021)
- [Semi-supervised Domain Adaptation based on Dual-level Domain Mixing for Semantic Segmentation](https://mlsanity.com/paper/2103.04705/) (2021)
- [LabOR: Labeling Only if Required for Domain Adaptive Semantic Segmentation](https://mlsanity.com/paper/2108.05570/) (2021)
- [Domain Adaptation for Semantic Segmentation via Patch-Wise Contrastive Learning](https://mlsanity.com/paper/2104.11056/) (2021)
- [Survey on Unsupervised Domain Adaptation for Semantic Segmentation for Visual Perception in Automated Driving](https://mlsanity.com/paper/2304.11928/) (2023)
- [Semantic Image Segmentation: Two Decades of Research](https://mlsanity.com/paper/2302.06378/) (2023)
- [Improving Semi-Supervised and Domain-Adaptive Semantic Segmentation with Self-Supervised Depth Estimation](https://mlsanity.com/paper/2108.12545/) (2021)
- [ACT: Semi-supervised Domain-adaptive Medical Image Segmentation with Asymmetric Co-training](https://mlsanity.com/paper/2206.02288/) (2022)
- [Unsupervised Domain Adaptation for Semantic Image Segmentation: a Comprehensive Survey](https://mlsanity.com/paper/2112.03241/) (2021)
- [SALAD: Source-free Active Label-Agnostic Domain Adaptation for Classification, Segmentation and Detection](https://mlsanity.com/paper/2205.12840/) (2022)
- [Deep Transfer Learning for Intelligent Vehicle Perception: a Survey](https://mlsanity.com/paper/2306.15110/) (2023)
- [Dual-Teacher++: Exploiting Intra-domain and Inter-domain Knowledge with Reliable Transfer for Cardiac Segmentation](https://mlsanity.com/paper/2101.02375/) (2021)
- [Co-Training for Unsupervised Domain Adaptation of Semantic Segmentation Models](https://mlsanity.com/paper/2205.15781/) (2022)
- [TrainSim: A Railway Simulation Framework for LiDAR and Camera Dataset Generation](https://mlsanity.com/paper/2302.14486/) (2023)
- [Semi-supervised Domain Adaptive Medical Image Segmentation through Consistency Regularized Disentangled Contrastive Learning](https://mlsanity.com/paper/2307.02798/) (2023)
- [Towards 3D Scene Understanding by Referring Synthetic Models](https://mlsanity.com/paper/2203.10546/) (2022)
- [Synthetic training data generation for deep learning based quality inspection](https://mlsanity.com/paper/2104.02980/) (2021)
- [TACS: Taxonomy Adaptive Cross-Domain Semantic Segmentation](https://mlsanity.com/paper/2109.04813/) (2021)
- [D2ADA: Dynamic Density-aware Active Domain Adaptation for Semantic Segmentation](https://mlsanity.com/paper/2202.06484/) (2022)
- [Active learning using weakly supervised signals for quality inspection](https://mlsanity.com/paper/2104.02973/) (2021)
- [Semi-supervised Domain Adaptation for Semantic Segmentation](https://mlsanity.com/paper/2110.10639/) (2021)
- [Semi-supervised domain adaptation with CycleGAN guided by a downstream task loss](https://mlsanity.com/paper/2208.08815/) (2022)
- [Model2Scene: Learning 3D Scene Representation via Contrastive Language-CAD Models Pre-training](https://mlsanity.com/paper/2309.16956/) (2023)
- [Bridging Distribution Shift and AI Safety: Conceptual and Methodological Synergies](https://mlsanity.com/paper/2505.22829/) (2025)
- [Active Domain Adaptation with Multi-level Contrastive Units for Semantic Segmentation](https://mlsanity.com/paper/2205.11192/) (2022)
- [IIDM: Inter and Intra-domain Mixing for Semi-supervised Domain Adaptation in Semantic Segmentation](https://mlsanity.com/paper/2308.15855/) (2023)
- [Learning Conditional Invariances through Non-Commutativity](https://mlsanity.com/paper/2402.11682/) (2024)
- [Improving generalization with synthetic training data for deep learning based quality inspection](https://mlsanity.com/paper/2202.12818/) (2022)

## Abstract (arXiv)

Learning segmentation from synthetic data and adapting to real data can significantly relieve human efforts in labelling pixel-level masks. A key challenge of this task is how to alleviate the data distribution discrepancy between the source and target domains, i.e. reducing domain shift. The common approach to this problem is to minimize the discrepancy between feature distributions from different domains through adversarial training. However, directly aligning the feature distribution globally cannot guarantee consistency from a local view (i.e. semantic-level), which prevents certain semantic knowledge learned on the source domain from being applied to the target domain. To tackle this issue, we propose a semi-supervised approach named Alleviating Semantic-level Shift (ASS), which can successfully promote the distribution consistency from both global and local views. Specifically, leveraging a small number of labeled data from the target domain, we directly extract semantic-level feature representations from both the source and the target domains by averaging the features corresponding to same categories advised by pixel-level masks. We then feed the produced features to the discriminator to conduct semantic-level adversarial learning, which collaborates with the adversarial learning from the global view to better alleviate the domain shift. We apply our ASS to two domain adaptation tasks, from GTA5 to Cityscapes and from Synthia to Cityscapes. Extensive experiments demonstrate that: (1) ASS can significantly outperform the current unsupervised state-of-the-arts by employing a small number of annotated samples from the target domain; (2) ASS can beat the oracle model trained on the whole target dataset by over 3 points by augmenting the synthetic source data with annotated samples from the target domain without suffering from the prevalent problem of overfitting to the source domain.