BMD: A General Class-balanced Multicentric Dynamic Prototype Strategy
for Source-free Domain Adaptation
- URL: http://arxiv.org/abs/2204.02811v1
- Date: Wed, 6 Apr 2022 13:23:02 GMT
- Title: BMD: A General Class-balanced Multicentric Dynamic Prototype Strategy
for Source-free Domain Adaptation
- Authors: Sanqing Qu, Guang Chen, Jing Zhang, Zhijun Li, Wei He, Dacheng Tao
- Abstract summary: Source-free Domain Adaptation (SFDA) aims to adapt a pre-trained source model to the unlabeled target domain without accessing the well-labeled source data.
To make up for the absence of source data, most existing methods introduced feature prototype based pseudo-labeling strategies.
We propose a general class-Balanced Multicentric Dynamic prototype strategy for the SFDA task.
- Score: 74.93176783541332
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Source-free Domain Adaptation (SFDA) aims to adapt a pre-trained source model
to the unlabeled target domain without accessing the well-labeled source data,
which is a much more practical setting due to the data privacy, security, and
transmission issues. To make up for the absence of source data, most existing
methods introduced feature prototype based pseudo-labeling strategies to
realize self-training model adaptation. However, feature prototypes are
obtained by instance-level predictions based feature clustering, which is
category-biased and tends to result in noisy labels since the visual domain
gaps between source and target are usually different between categories. In
addition, we found that a monocentric feature prototype may be ineffective to
represent each category and introduce negative transfer, especially for those
hard-transfer data. To address these issues, we propose a general
class-Balanced Multicentric Dynamic prototype (BMD) strategy for the SFDA task.
Specifically, for each target category, we first introduce a global inter-class
balanced sampling strategy to aggregate potential representative target
samples. Then, we design an intra-class multicentric clustering strategy to
achieve more robust and representative prototypes generation. In contrast to
existing strategies that update the pseudo label at a fixed training period, we
further introduce a dynamic pseudo labeling strategy to incorporate network
update information during model adaptation. Extensive experiments show that the
proposed model-agnostic BMD strategy significantly improves representative SFDA
methods to yield new state-of-the-art results, e.g., improving SHOT from 82.9\%
to 85.8\% on VisDA-C and NRC from 52.6\% to 57.0\% on PointDA. The code is
available at https://github.com/ispc-lab/BMD.
Related papers
- ProtoGMM: Multi-prototype Gaussian-Mixture-based Domain Adaptation Model for Semantic Segmentation [0.8213829427624407]
Domain adaptive semantic segmentation aims to generate accurate and dense predictions for an unlabeled target domain.
We propose the ProtoGMM model, which incorporates the GMM into contrastive losses to perform guided contrastive learning.
To achieve increased intra-class semantic similarity, decreased inter-class similarity, and domain alignment between the source and target domains, we employ multi-prototype contrastive learning.
arXiv Detail & Related papers (2024-06-27T14:50:50Z) - Consistency Regularization for Generalizable Source-free Domain
Adaptation [62.654883736925456]
Source-free domain adaptation (SFDA) aims to adapt a well-trained source model to an unlabelled target domain without accessing the source dataset.
Existing SFDA methods ONLY assess their adapted models on the target training set, neglecting the data from unseen but identically distributed testing sets.
We propose a consistency regularization framework to develop a more generalizable SFDA method.
arXiv Detail & Related papers (2023-08-03T07:45:53Z) - Semi-supervised Domain Adaptation via Prototype-based Multi-level
Learning [4.232614032390374]
In semi-supervised domain adaptation (SSDA), a few labeled target samples of each class help the model to transfer knowledge representation from the fully labeled source domain to the target domain.
We propose a Prototype-based Multi-level Learning (ProML) framework to better tap the potential of labeled target samples.
arXiv Detail & Related papers (2023-05-04T10:09:30Z) - Upcycling Models under Domain and Category Shift [95.22147885947732]
We introduce an innovative global and local clustering learning technique (GLC)
We design a novel, adaptive one-vs-all global clustering algorithm to achieve the distinction across different target classes.
Remarkably, in the most challenging open-partial-set DA scenario, GLC outperforms UMAD by 14.8% on the VisDA benchmark.
arXiv Detail & Related papers (2023-03-13T13:44:04Z) - A Prototype-Oriented Clustering for Domain Shift with Source Privacy [66.67700676888629]
We introduce Prototype-oriented Clustering with Distillation (PCD) to improve the performance and applicability of existing methods.
PCD first constructs a source clustering model by aligning the distributions of prototypes and data.
It then distills the knowledge to the target model through cluster labels provided by the source model while simultaneously clustering the target data.
arXiv Detail & Related papers (2023-02-08T00:15:35Z) - Polycentric Clustering and Structural Regularization for Source-free
Unsupervised Domain Adaptation [20.952542421577487]
Source-Free Domain Adaptation (SFDA) aims to solve the domain adaptation problem by transferring the knowledge learned from a pre-trained source model to an unseen target domain.
Most existing methods assign pseudo-labels to the target data by generating feature prototypes.
In this paper, a novel framework named PCSR is proposed to tackle SFDA via a novel intra-class Polycentric Clustering and Structural Regularization strategy.
arXiv Detail & Related papers (2022-10-14T02:20:48Z) - Source-free Domain Adaptation via Avatar Prototype Generation and
Adaptation [34.45208248728318]
We study a practical domain adaptation task in which we cannot access source domain data due to data privacy issues.
The lack of source data and target domain labels makes model adaptation very challenging.
We propose a Contrastive Prototype Generation and Adaptation (CPGA) method to exploit hidden knowledge in the source model.
arXiv Detail & Related papers (2021-06-18T08:30:54Z) - Semi-Supervised Domain Adaptation with Prototypical Alignment and
Consistency Learning [86.6929930921905]
This paper studies how much it can help address domain shifts if we further have a few target samples labeled.
To explore the full potential of landmarks, we incorporate a prototypical alignment (PA) module which calculates a target prototype for each class from the landmarks.
Specifically, we severely perturb the labeled images, making PA non-trivial to achieve and thus promoting model generalizability.
arXiv Detail & Related papers (2021-04-19T08:46:08Z) - OVANet: One-vs-All Network for Universal Domain Adaptation [78.86047802107025]
Existing methods manually set a threshold to reject unknown samples based on validation or a pre-defined ratio of unknown samples.
We propose a method to learn the threshold using source samples and to adapt it to the target domain.
Our idea is that a minimum inter-class distance in the source domain should be a good threshold to decide between known or unknown in the target.
arXiv Detail & Related papers (2021-04-07T18:36:31Z)
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.