mlsanityData to 14 Oct 2026

Continual Adaptation of Visual Representations via Domain Randomization and Meta-learning

arXiv 2012.04324 ↗First posted 8 Dec 2020cs.CV cs.AI cs.LGcited by 55

Not on the map. Why?

Abstract (arXiv)

Most standard learning approaches lead to fragile models which are prone to drift when sequentially trained on samples of a different nature - the well-known "catastrophic forgetting" issue. In particular, when a model consecutively learns from different visual domains, it tends to forget the past domains in favor of the most recent ones. In this context, we show that one way to learn models that are inherently more robust against forgetting is domain randomization - for vision tasks, randomizing the current domain's distribution with heavy image manipulations. Building on this result, we devise a meta-learning strategy where a regularizer explicitly penalizes any loss associated with transferring the model from the current domain to different "auxiliary" meta-domains, while also easing adaptation to them. Such meta-domains are also generated through randomized image manipulations. We empirically demonstrate in a variety of experiments - spanning from classification to semantic segmentation - that our approach results in models that are less prone to catastrophic forgetting when transferred to new domains.

Key facts. Cited by 55 papers in the map (data through 14 Oct 2026). First posted on arXiv 8 Dec 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 3; 2022: 14; 2023: 18; 2024: 13; 2025: 3; 2026: 4.

55 papers.

Sort:

By area:

  1. 2021Domain Generalization: A Surveycited by 764
  2. 2022Continual Test-Time Domain Adaptationcited by 525
  3. 2023Conditional Image-to-Video Generation with Latent Flow Diffusion Modelscited by 181
  4. 2022Fake it till you make it: Learning transferable representations from synthetic ImageNet clonescited by 172
  5. 2022SHIFT: A Synthetic Driving Dataset for Continuous Multi-Task Domain Adaptationcited by 146
  6. 2023A Comprehensive Survey of Forgetting in Deep Learning Beyond Continual Learningcited by 69
  7. 2021Recent Advances of Continual Learning in Computer Vision: An Overviewcited by 55
  8. 2023Synthetic Datasets for Autonomous Driving: A Surveycited by 48
  9. 2024Online DPO: Online Direct Preference Optimization with Fast-Slow Chasingcited by 39
  10. 2023Collaborating Foundation Models for Domain Generalized Semantic Segmentationcited by 34
  11. 2023Semantic Image Segmentation: Two Decades of Researchcited by 27
  12. 2022General Incremental Learning with Domain-aware Categorical Representationscited by 26
38 more
  1. 2022On Generalizing Beyond Domains in Cross-Domain Continual Learningcited by 25
  2. 2022On the Road to Online Adaptation for Semantic Image Segmentationcited by 23
  3. 2022InCloud: Incremental Learning for Point Cloud Place Recognitioncited by 22
  4. 2023Does Continual Learning Equally Forget All Parameters?cited by 18
  5. 2022Cloud-Device Collaborative Adaptation to Continual Changing Environments in the Real-worldcited by 17
  6. 2024Parameter Efficient Fine-tuning of Self-supervised ViTs without Catastrophic Forgettingcited by 17
  7. 2023Fairness Continual Learning Approach to Semantic Scene Understanding in Open-World Environmentscited by 16
  8. 2022Label-Efficient Domain Generalization via Collaborative Exploration and Generalizationcited by 15
  9. 2023FALCON: Fairness Learning via Contrastive Attention Approach to Continual Semantic Scene Understandingcited by 13
  10. 2022Multi-Head Distillation for Continual Unsupervised Domain Adaptation in Semantic Segmentationcited by 12
  11. 2023When Meta-Learning Meets Online and Continual Learning: A Surveycited by 12
  12. 2022CONDA: Continual Unsupervised Domain Adaptation Learning in Visual Perception for Self-Driving Carscited by 11
  13. 2023Cloud-Device Collaborative Learning for Multimodal Large Language Modelscited by 11
  14. 2022Learning with Style: Continual Semantic Segmentation Across Tasks and Domainscited by 9
  15. 2023Exploring Compositional Visual Generation with Latent Classifier Guidancecited by 9
  16. 2024Versatile Incremental Learning: Towards Class and Domain-Agnostic Incremental Learningcited by 9
  17. 2023Assessor-Guided Learning for Continual Environmentscited by 7
  18. 2023Algorithmic Fairness Generalization under Covariate and Dependence Shifts Simultaneouslycited by 7
  19. 2023Open-World Pose Transfer via Sequential Test-Time Adaptioncited by 4
  20. 2024Continual Domain Adversarial Adaptation via Double-Head Discriminatorscited by 4
  21. 2024FEED: Fairness-Enhanced Meta-Learning for Domain Generalizationcited by 3
  22. 2021Center Loss Regularization for Continual Learningcited by 2
  23. 2022LLEDA -- Lifelong Self-Supervised Domain Adaptationcited by 2
  24. 2023Lifelong Change Detection: Continuous Domain Adaptation for Small Object Change Detection in Every Robot Navigationcited by 2
  25. 2023Multivariate Prototype Representation for Domain-Generalized Incremental Learningcited by 2
  26. 2024Using Diffusion Models as Generative Replay in Continual Federated Learning -- What will Happen?cited by 2
  27. 2023Confidence Attention and Generalization Enhanced Distillation for Continuous Video Domain Adaptationcited by 1
  28. 2023Continual Source-Free Unsupervised Domain Adaptationcited by 1
  29. 2024Weatherproofing Retrieval for Localization with Generative AI and Geometric Consistencycited by 1
  30. 2024Cross-Domain Continual Learning via CLAMPcited by 1
  31. 2024Adaptive Cascading Network for Continual Test-Time Adaptationcited by 1
  32. 2024Domains as Objectives: Domain-Uncertainty-Aware Policy Optimization through Explicit Multi-Domain Convex Coverage Set Learningcited by 1
  33. 2024Improving Object Detection via Local-global Contrastive Learningcited by 1
  34. 2024SeFENet: Robust Deep Homography Estimation via Semantic-Driven Feature Enhancementcited by 1
  35. 2022Progressive Voronoi Diagram Subdivision: Towards A Holistic Geometric Framework for Exemplar-free Class-Incremental Learningcited by 0
  36. 2024Incremental Open-set Domain Adaptationcited by 0
  37. 2025Universal Incremental Learning: Mitigating Confusion from Inter- and Intra-task Distribution Randomnesscited by 0
  38. 2025Incremental Object Keypoint Learningcited by 0

50 most cited of 55 All 55 (JSON)