mlsanityData to 7 Oct 2026

PASTA: Proportional Amplitude Spectrum Training Augmentation for Syn-to-Real Domain Generalization

arXiv 2212.00979 ↗First posted 2 Dec 2022cs.CV cs.AI cs.LGcited by 38

Not on the map. Why?

Abstract (arXiv)

Synthetic data offers the promise of cheap and bountiful training data for settings where labeled real-world data is scarce. However, models trained on synthetic data significantly underperform when evaluated on real-world data. In this paper, we propose Proportional Amplitude Spectrum Training Augmentation (PASTA), a simple and effective augmentation strategy to improve out-of-the-box synthetic-to-real (syn-to-real) generalization performance. PASTA perturbs the amplitude spectra of synthetic images in the Fourier domain to generate augmented views. Specifically, with PASTA we propose a structured perturbation strategy where high-frequency components are perturbed relatively more than the low-frequency ones. For the tasks of semantic segmentation (GTAV-to-Real), object detection (Sim10K-to-Real), and object recognition (VisDA-C Syn-to-Real), across a total of 5 syn-to-real shifts, we find that PASTA outperforms more complex state-of-the-art generalization methods while being complementary to the same.

Key facts. Cited by 38 papers in the map (data through 7 Oct 2026). Builds on 68 of its 91 references that are in the map. First posted on arXiv 2 Dec 2022.

Builds on

68 of 91 references.

References · 68

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  4. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  5. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  6. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  7. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  8. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
  9. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  10. 2018Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentationcited by 3,210
  11. 2021SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformerscited by 2,923
  12. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
38 more
  1. 2019RandAugment: Practical automated data augmentation with a reduced search spacecited by 2,261
  2. 2019Invariant Risk Minimizationcited by 1,832
  3. 2019MMDetection: Open MMLab Detection Toolbox and Benchmarkcited by 1,379
  4. 2018BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learningcited by 1,282
  5. 2020In Search of Lost Domain Generalizationcited by 1,042
  6. 2017Deeper, Broader and Artier Domain Generalizationcited by 983
  7. 2016Playing for Data: Ground Truth from Computer Gamescited by 816
  8. 2017Learning to Generalize: Meta-Learning for Domain Generalizationcited by 762
  9. 2013Domain Generalization via Invariant Feature Representationcited by 622
  10. 2021Domain Generalization with MixStylecited by 563
  11. 2020FDA: Fourier Domain Adaptation for Semantic Segmentationcited by 540
  12. 2018Domain Adaptive Faster R-CNN for Object Detection in the Wildcited by 442
  13. 2017VisDA: The Visual Domain Adaptation Challengecited by 432
  14. 2021SWAD: Domain Generalization by Seeking Flat Minimacited by 431
  15. 2019Domain Generalization via Model-Agnostic Learning of Semantic Featurescited by 405
  16. 2018Two at Once: Enhancing Learning and Generalization Capacities via IBN-Netcited by 361
  17. 2021DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentationcited by 299
  18. 2020Learning to Learn Single Domain Generalizationcited by 295
  19. 2019A Fourier Perspective on Model Robustness in Computer Visioncited by 288
  20. 2021A Fourier-based Framework for Domain Generalizationcited by 287
  21. 2018Strong-Weak Distribution Alignment for Adaptive Object Detectioncited by 269
  22. 2021Gradient Matching for Domain Generalizationcited by 258
  23. 2022Contrastive Test-Time Adaptationcited by 258
  24. 2015Domain Generalization for Object Recognition with Multi-task Autoencoderscited by 253
  25. 2019Domain Randomization and Pyramid Consistency: Simulation-to-Real Generalization without Accessing Target Domain Datacited by 245
  26. 2021RobustNet: Improving Domain Generalization in Urban-Scene Segmentation via Instance Selective Whiteningcited by 213
  27. 2016Driving in the Matrix: Can Virtual Worlds Replace Human-Generated Annotations for Real World Tasks?cited by 194
  28. 2017Generate To Adapt: Aligning Domains using Generative Adversarial Networkscited by 183
  29. 2015ParseNet: Looking Wider to See Bettercited by 176
  30. 2021Learning to Diversify for Single Domain Generalizationcited by 170
  31. 2021FSDR: Frequency Space Domain Randomization for Domain Generalizationcited by 163
  32. 2022HRDA: Context-Aware High-Resolution Domain-Adaptive Semantic Segmentationcited by 163
  33. 2020Learning to Balance Specificity and Invariance for In and Out of Domain Generalizationcited by 128
  34. 2020Harmonizing Transferability and Discriminability for Adapting Object Detectorscited by 125
  35. 2022Uncertainty Modeling for Out-of-Distribution Generalizationcited by 122
  36. 2017Adversarial Dropout Regularizationcited by 106
  37. 2020Every Pixel Matters: Center-aware Feature Alignment for Domain Adaptive Object Detectorcited by 106
  38. 2021MeGA-CDA: Memory Guided Attention for Category-Aware Unsupervised Domain Adaptive Object Detectioncited by 106

50 most cited of 68 All 68 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 4; 2024: 10; 2025: 17; 2026: 7.

38 papers.

Sort:

By area:

  1. 2023Stronger, Fewer, & Superior: Harnessing Vision Foundation Models for Domain Generalized Semantic Segmentationcited by 82
  2. 2025Mamba as a Bridge: Where Vision Foundation Models Meet Vision Language Models for Domain-Generalized Semantic Segmentationcited by 23
  3. 2024SoMA: Singular Value Decomposed Minor Components Adaptation for Domain Generalizable Representation Learningcited by 22
  4. 2023Strong but simple: A Baseline for Domain Generalized Dense Perception by CLIP-based Transfer Learningcited by 20
  5. 2025FisherTune: Fisher-Guided Robust Tuning of Vision Foundation Models for Domain Generalized Segmentationcited by 15
  6. 2023Bridging the Sim2Real gap with CARE: Supervised Detection Adaptation with Conditional Alignment and Reweightingcited by 9
  7. 2024Exploring Semantic Consistency and Style Diversity for Domain Generalized Semantic Segmentationcited by 7
  8. 2024Medical Image Segmentation via Single-Source Domain Generalization with Random Amplitude Spectrum Synthesiscited by 6
  9. 2025Stronger, Steadier & Superior: Geometric Consistency in Depth VFM Forges Domain Generalized Semantic Segmentationcited by 6
  10. 2024Syn2Real Domain Generalization for Underwater Mine-like Object Detection Using Side-Scan Sonarcited by 4
  11. 2025Exploiting Domain Properties in Language-Driven Domain Generalization for Semantic Segmentationcited by 4
  12. 2024Synthetica: Large Scale Synthetic Data for Robot Perceptioncited by 3
26 more
  1. 2025Generalizable Multispectral Land Cover Classification via Frequency-Aware Mixture of Low-Rank Token Expertscited by 3
  2. 2025Bridging Annotation Gaps: Transferring Labels to Align Object Detection Datasetscited by 3
  3. 2024MergeOcc: Bridge the Domain Gap between Different LiDARs for Robust Occupancy Predictioncited by 2
  4. 2024FIESTA: Fourier-Based Semantic Augmentation with Uncertainty Guidance for Enhanced Domain Generalizability in Medical Image Segmentationcited by 2
  5. 2025Rein++: Efficient Generalization and Adaptation for Semantic Segmentation with Vision Foundation Modelscited by 2
  6. 2025Adaptive Frequency Domain Alignment Network for Medical image segmentationcited by 2
  7. 2026Open-Vocabulary Domain Generalization in Urban-Scene Segmentationcited by 2
  8. 2023AUGCAL: Improving Sim2Real Adaptation by Uncertainty Calibration on Augmented Synthetic Imagescited by 1
  9. 2024Semantic-Rearrangement-Based Multi-Level Alignment for Domain Generalized Segmentationcited by 1
  10. 2025What is the Added Value of UDA in the VFM Era?cited by 1
  11. 2025Leveraging Depth and Language for Open-Vocabulary Domain-Generalized Semantic Segmentationcited by 1
  12. 2026Local Precise Refinement: A Dual-Gated Mixture-of-Experts for Enhancing Foundation Model Generalization against Spectral Shiftscited by 1
  13. 2026PiLoT: Neural Pixel-to-3D Registration for UAV-based Ego and Target Geo-localizationcited by 1
  14. 2024OSSA: Unsupervised One-Shot Style Adaptationcited by 0
  15. 2024A Study on Unsupervised Domain Adaptation for Semantic Segmentation in the Era of Vision-Language Modelscited by 0
  16. 2025Casual Inference via Style Bias Deconfounding for Domain Generalizationcited by 0
  17. 2025Depth-Sensitive Soft Suppression with RGB-D Inter-Modal Stylization Flow for Domain Generalization Semantic Segmentationcited by 0
  18. 2025Frequency Composition for Compressed and Domain-Adaptive Neural Networkscited by 0
  19. 2025Set Pivot Learning: Redefining Generalized Segmentation with Vision Foundation Modelscited by 0
  20. 2025FOCUS: Frequency-Optimized Conditioning of DiffUSion Models for mitigating catastrophic forgetting during Test-Time Adaptationcited by 0
  21. 2025IELDG: Suppressing Domain-Specific Noise with Inverse Evolution Layers for Domain Generalized Semantic Segmentationcited by 0
  22. 2025Domain Generalization for Semantic Segmentation: A Surveycited by 0
  23. 2026RecycleLoRA: Rank-Revealing QR-Based Dual-LoRA Subspace Adaptation for Domain Generalized Semantic Segmentationcited by 0
  24. 2026LASA: Language-and-Source-Anchored Alignment for Domain Generalized Semantic Segmentationcited by 0
  25. 2026Hierarchical Prompt Injector for Domain Generalization Segmentationcited by 0
  26. 2026Domain generalization and synthetic data in object detection: the enabler, the probe, and the gapcited by 0