mlsanityData to 7 Oct 2026

Fuzzy-aware Loss for Source-free Domain Adaptation in Visual Emotion Recognition

arXiv 2501.15519 ↗First posted 26 Jan 2025cs.CV cs.LGcited by 3

Not on the map. Why?

Abstract (arXiv)

Source-free domain adaptation in visual emotion recognition (SFDA-VER) is a highly challenging task that requires adapting VER models to the target domain without relying on source data, which is of great significance for data privacy protection. However, due to the unignorable disparities between visual emotion data and traditional image classification data, existing SFDA methods perform poorly on this task. In this paper, we investigate the SFDA-VER task from a fuzzy perspective and identify two key issues: fuzzy emotion labels and fuzzy pseudo-labels. These issues arise from the inherent uncertainty of emotion annotations and the potential mispredictions in pseudo-labels. To address these issues, we propose a novel fuzzy-aware loss (FAL) to enable the VER model to better learn and adapt to new domains under fuzzy labels. Specifically, FAL modifies the standard cross entropy loss and focuses on adjusting the losses of non-predicted categories, which prevents a large number of uncertain or incorrect predictions from overwhelming the VER model during adaptation. In addition, we provide a theoretical analysis of FAL and prove its robustness in handling the noise in generated pseudo-labels. Extensive experiments on 26 domain adaptation sub-tasks across three benchmark datasets demonstrate the effectiveness of our method.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 34 of its 67 references that are in the map. First posted on arXiv 26 Jan 2025.

Builds on

34 of 67 references.

References · 34

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2018Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labelscited by 994
  4. 2019When Does Label Smoothing Help?cited by 985
  5. 2020Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptationcited by 844
  6. 2019Symmetric Cross Entropy for Robust Learning with Noisy Labelscited by 499
  7. 2017Robust Loss Functions under Label Noise for Deep Neural Networkscited by 441
  8. 2018Deep Facial Expression Recognition: A Surveycited by 269
  9. 2021Exploiting the Intrinsic Neighborhood Structure for Source-free Domain Adaptationcited by 176
  10. 2021Generalized Source-free Domain Adaptationcited by 175
  11. 2019NLNL: Negative Learning for Noisy Labelscited by 151
  12. 2021Source-free Domain Adaptation via Avatar Prototype Generation and Adaptationcited by 111
22 more
  1. 2023A Comprehensive Survey on Source-free Domain Adaptationcited by 104
  2. 2022Attracting and Dispersing: A Simple Approach for Source-free Domain Adaptationcited by 93
  3. 2023EmoSet: A Large-scale Visual Emotion Dataset with Rich Attributescited by 81
  4. 2020Context Based Emotion Recognition using EMOTIC Datasetcited by 73
  5. 2016Building a Large Scale Dataset for Image Emotion Recognition: The Fine Print and The Benchmarkcited by 71
  6. 2022PolyLoss: A Polynomial Expansion Perspective of Classification Loss Functionscited by 62
  7. 2019M3ER: Multiplicative Multimodal Emotion Recognition Using Facial, Textual, and Speech Cuescited by 56
  8. 2022Divide and Contrast: Source-free Domain Adaptation via Adaptive Contrastive Learningcited by 53
  9. 2021Affective Image Content Analysis: Two Decades Review and New Perspectivescited by 47
  10. 2015Robust Image Sentiment Analysis Using Progressively Trained and Domain Transferred Deep Networkscited by 46
  11. 2023When Source-Free Domain Adaptation Meets Learning with Noisy Labelscited by 46
  12. 2022Mitigating Memorization of Noisy Labels by Clipping the Model Predictioncited by 40
  13. 2019STEP: Spatial Temporal Graph Convolutional Networks for Emotion Perception from Gaitscited by 36
  14. 2023Context De-confounded Emotion Recognitioncited by 34
  15. 2024A Survey of Embodied Learning for Object-Centric Robotic Manipulationcited by 34
  16. 2022MDAN: Multi-level Dependent Attention Network for Visual Emotion Analysiscited by 29
  17. 2021SOLVER: Scene-Object Interrelated Visual Emotion Reasoning Networkcited by 18
  18. 2014Visual Sentiment Prediction with Deep Convolutional Neural Networkscited by 6
  19. 2019Deep Semantic Parsing of Freehand Sketches with Homogeneous Transformation, Soft-Weighted Loss, and Staged Learningcited by 5
  20. 2019Sketch-Specific Data Augmentation for Freehand Sketch Recognitioncited by 4
  21. 2022How Well Do Self-Supervised Methods Perform in Cross-Domain Few-Shot Learning?cited by 4
  22. 2024Optimized Gradient Clipping for Noisy Label Learningcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2025Personalized Feature Translation for Expression Recognition: An Efficient Source-Free Domain Adaptation Methodcited by 8
  2. 2026ProCal: Probability Calibration for Neighborhood-Guided Source-Free Domain Adaptationcited by 1
  3. 2026Evolution-Inspired Sample Competition for Deep Neural Network Optimizationcited by 0