• Home
  • Search
  • DI2SDiff++: Activity Style Decomposition and Diffusion-Based Fusion for Cross-Person Generalization in Activity Recognition
  • Cite Icon1
  • https://doi.org/10.1109/tmc.2025.3572220Copy DOI Icon

DI2SDiff++: Activity Style Decomposition and Diffusion-Based Fusion for Cross-Person Generalization in Activity Recognition

Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Existing domain generalization (DG) methods for cross-person sensor-based activity recognition tasks often struggle to capture both intra- and inter-domain style diversity, leading to significant domain gaps with the target domain. In this study, we explore a novel perspective to tackle this problem, a process conceptualized as domain padding. This proposal aims to enrich the domain diversity by synthesizing intra- and inter-domain style data while maintaining robustness to class labels. We instantiate this concept using a conditional diffusion model and introduce a style-fused sampling strategy to enhance data generation diversity, termed Diversified Intra- and Inter-domain distributions via activity Style-fused Diffusion modeling (DI2SDiff). In contrast to traditional condition-guided sampling, our style-fused sampling strategy allows for the flexible use of one or more random style representations from the same class to guide data synthesis. This feature presents a notable advancement: it allows for the maximum utilization of possible combinations among existing styles to generate a broad spectrum of new style instances. We further extend DI2SDiff into DI2SDiff++ by enhancing the diversity of style guidance. Specifically, DI2SDiff++ integrates a multi-head style conditioner to provide multiple distinct, decomposed substyles and introduces a substyle-fused sampling strategy that allows cross-class substyle fusion for broader guidance. Empirical evaluations on a wide range of datasets demonstrate that our generated data achieves remarkable diversity within the domain space. Both intra- and inter-domain generated data have been proven significant and valuable, enabling DI2SDiff and DI2SDiff++ to surpass state-of-the-art DG methods in various cross-person activity recognition tasks.

Similar Papers
  • Conference Article
  • Citations9

Diverse Intra- and Inter-Domain Activity Style Fusion for Cross-Person Generalization in Activity Recognition

  • Aug 24, 2024
  • Junru Zhang +6
  • Book Chapter
  • Citations31

Self-distilled Vision Transformer for Domain Generalization

  • Jan 01, 2023
  • Maryam Sultana +4
  • Research Article
  • Citations25

Domain Generalization for Face Anti-Spoofing via Negative Data Augmentation

  • Jan 01, 2023
  • IEEE Transactions on Information Forensics and Security
  • Weihang Wang +4
  • Research Article
  • Citations2

Multisource Domain Separation Network for Industrial Intelligent Monitoring in Unseen Conditions.

  • Jan 01, 2025
  • IEEE transactions on cybernetics
  • Zidi Jia +3
  • Dissertation
  • Citations1

Domain adaptation and generalization for visual recognition

  • Jan 01, 2017
  • Li Niu
  • Research Article
  • Citations11

Learning Category-Invariant Disentangled Features for Domain Generalization in Machine Fault Diagnosis

  • Jan 01, 2026
  • IEEE/ASME Transactions on Mechatronics
  • Zuoyi Chen +3
  • Research Article
  • Citations1

Single-Source Domain Defect-Aware Adaptation and Style-Modulated Generalization Network for Multispectral Image Segmentation.

  • Mar 01, 2026
  • IEEE transactions on cybernetics
  • Wei Li +4
  • Conference Article
  • Citations1436

Deeper, Broader and Artier Domain Generalization

  • Oct 01, 2017
  • Da Li +3
  • Conference Article
  • Citations51

Federated Domain Generalization for Image Recognition via Cross-Client Style Transfer

  • Jan 01, 2023
  • Junming Chen +3
  • Conference Article
  • Citations5

Simple Domain Generalization Methods are Strong Baselines for Open Domain Generalization

  • Jun 30, 2024
  • Masashi Noguchi +1
  • Research Article
  • Citations13

PLACE Dropout: A Progressive Layer-wise and Channel-wise Dropout for Domain Generalization

  • Oct 23, 2023
  • ACM Transactions on Multimedia Computing, Communications, and Applications
  • Jintao Guo +3
  • Research Article
  • Citations8

Unsupervised Contrastive Learning-Based Single Domain Generalization Method for Intelligent Bearing Fault Diagnosis

  • Jan 15, 2025
  • IEEE Sensors Journal
  • Qiang Wu +4
  • Book Chapter
  • Citations7

Generalizing Nucleus Recognition Model in Multi-source Ki67 Immunohistochemistry Stained Images via Domain-Specific Pruning

  • Jan 01, 2021
  • Jiatong Cai +6
  • Research Article
  • Citations15

Prompt-Driven Latent Domain Generalization for Medical Image Classification.

  • Jan 01, 2025
  • IEEE transactions on medical imaging
  • Siyuan Yan +9
  • Research Article
  • Citations101

Self-Domain Adaptation for Face Anti-Spoofing

  • May 18, 2021
  • Proceedings of the AAAI Conference on Artificial Intelligence
  • Jingjing Wang +5
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.