Attention Diversification for Domain Generalization

• URL: http://arxiv.org/abs/2210.04206v1
• Date: Sun, 9 Oct 2022 09:15:21 GMT
• Title: Attention Diversification for Domain Generalization
• Authors: Rang Meng, Xianfeng Li, Weijie Chen, Shicai Yang, Jie Song, Xinchao Wang, Lei Zhang, Mingli Song, Di Xie, and Shiliang Pu
• Abstract summary: Convolutional neural networks (CNNs) have demonstrated gratifying results at learning discriminative features. When applied to unseen domains, state-of-the-art models are usually prone to errors due to domain shift. We propose a novel Attention Diversification framework, in which Intra-Model and Inter-Model Attention Diversification Regularization are collaborated.
• Score: 92.02038576148774
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Convolutional neural networks (CNNs) have demonstrated gratifying results at learning discriminative features. However, when applied to unseen domains, state-of-the-art models are usually prone to errors due to domain shift. After investigating this issue from the perspective of shortcut learning, we find the devils lie in the fact that models trained on different domains merely bias to different domain-specific features yet overlook diverse task-related features. Under this guidance, a novel Attention Diversification framework is proposed, in which Intra-Model and Inter-Model Attention Diversification Regularization are collaborated to reassign appropriate attention to diverse task-related features. Briefly, Intra-Model Attention Diversification Regularization is equipped on the high-level feature maps to achieve in-channel discrimination and cross-channel diversification via forcing different channels to pay their most salient attention to different spatial locations. Besides, Inter-Model Attention Diversification Regularization is proposed to further provide task-related attention diversification and domain-related attention suppression, which is a paradigm of "simulate, divide and assemble": simulate domain shift via exploiting multiple domain-specific models, divide attention maps into task-related and domain-related groups, and assemble them within each group respectively to execute regularization. Extensive experiments and analyses are conducted on various benchmarks to demonstrate that our method achieves state-of-the-art performance over other competing methods. Code is available at https://github.com/hikvision-research/DomainGeneralization.

Related papers

• Generalize or Detect? Towards Robust Semantic Segmentation Under Multiple Distribution Shifts [56.57141696245328]
  In open-world scenarios, where both novel classes and domains may exist, an ideal segmentation model should detect anomaly classes for safety. Existing methods often struggle to distinguish between domain-level and semantic-level distribution shifts.
  arXiv  Detail & Related papers  (2024-11-06T11:03:02Z)
• Towards Domain-Specific Features Disentanglement for Domain Generalization [23.13095840134744]
  We propose a novel contrastive-based disentanglement method CDDG to exploit the over-looked domain-specific features. Specifically, CDDG learns to decouple inherent mutually exclusive features by leveraging them in the latent space. Experiments conducted on various benchmark datasets demonstrate the superiority of our method compared to other state-of-the-art approaches.
  arXiv  Detail & Related papers  (2023-10-04T17:51:02Z)
• Multi-Head Distillation for Continual Unsupervised Domain Adaptation in Semantic Segmentation [38.10483890861357]
  This work focuses on a novel framework for learning UDA, continuous UDA, in which models operate on multiple target domains discovered sequentially. We propose MuHDi, for Multi-Head Distillation, a method that solves the catastrophic forgetting problem, inherent in continual learning tasks.
  arXiv  Detail & Related papers  (2022-04-25T14:03:09Z)
• Label Distribution Learning for Generalizable Multi-source Person Re-identification [48.77206888171507]
  Person re-identification (Re-ID) is a critical technique in the video surveillance system. It is difficult to directly apply the supervised model to arbitrary unseen domains. We propose a novel label distribution learning (LDL) method to address the generalizable multi-source person Re-ID task.
  arXiv  Detail & Related papers  (2022-04-12T15:59:10Z)
• CADG: A Model Based on Cross Attention for Domain Generalization [6.136770353307872]
  In Domain Generalization (DG) tasks, models are trained by using only training data from the source domains to achieve generalization on an unseen target domain. We design a model named CADG (cross attention for domain generalization), wherein cross attention plays a important role, to address distribution shift problem. Experiments show that our proposed method achieves state-of-the-art performance on a variety of domain generalization benchmarks.
  arXiv  Detail & Related papers  (2022-03-31T14:35:21Z)
• AFAN: Augmented Feature Alignment Network for Cross-Domain Object Detection [90.18752912204778]
  Unsupervised domain adaptation for object detection is a challenging problem with many real-world applications. We propose a novel augmented feature alignment network (AFAN) which integrates intermediate domain image generation and domain-adversarial training. Our approach significantly outperforms the state-of-the-art methods on standard benchmarks for both similar and dissimilar domain adaptations.
  arXiv  Detail & Related papers  (2021-06-10T05:01:20Z)
• Contrastive ACE: Domain Generalization Through Alignment of Causal Mechanisms [34.99779761100095]
  Domain generalization aims to learn knowledge invariant across different distributions. We consider the causal invariance of the average causal effect of the features to the labels.
  arXiv  Detail & Related papers  (2021-06-02T04:01:22Z)
• Channel-wise Alignment for Adaptive Object Detection [66.76486843397267]
  Generic object detection has been immensely promoted by the development of deep convolutional neural networks. Existing methods on this task usually draw attention on the high-level alignment based on the whole image or object of interest. In this paper, we realize adaptation from a thoroughly different perspective, i.e., channel-wise alignment.
  arXiv  Detail & Related papers  (2020-09-07T02:42:18Z)
• Learning to Combine: Knowledge Aggregation for Multi-Source Domain Adaptation [56.694330303488435]
  We propose a Learning to Combine for Multi-Source Domain Adaptation (LtC-MSDA) framework. In the nutshell, a knowledge graph is constructed on the prototypes of various domains to realize the information propagation among semantically adjacent representations. Our approach outperforms existing methods with a remarkable margin.
  arXiv  Detail & Related papers  (2020-07-17T07:52:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.