Unsupervised Domain Adaption Harnessing Vision-Language Pre-Training

作     者：Zhou, Wenlve Zhou, Zhiheng 

作者机构：South China Univ Technol Sch Elect & Informat Engn Guangzhou 510641 Guangdong Peoples R China South China Univ Technol Minist Educ Key Lab Big Data & Intelligent Robot Guangzhou 510641 Guangdong Peoples R China 

出 版 物：《IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY》 (IEEE Trans Circuits Syst Video Technol)

年 卷 期：2024年第34卷第9期

页      面：8201-8214页

核心收录：

学科分类：0808[工学-电气工程] 08[工学] 

基　　金：National Key Research and Development Program of China [2022YFF0607001] Guangdong Basic and Applied Basic Research Foundation [2023A1515010993] Guangdong Provincial Key Laboratory of Human Digital Twin [2022B1212010004] Guangzhou City Science and Technology Research Projects [2023B01J0011] Jiangmen Science and Technology Research Projects Shaoguan Science and Technology Research Project Foshan Science and Technology Research Project 

主　　题：Unsupervised domain adaptation vision-language pre-training cross-modal knowledge distillation residual sparse training model deployment 

摘      要：This paper addresses two vital challenges in Unsupervised Domain Adaptation (UDA) with a focus on harnessing the power of Vision-Language Pre-training (VLP) models. Firstly, UDA has primarily relied on ImageNet pre-trained models. However, the potential of VLP models in UDA remains largely unexplored. The rich representation of VLP models holds significant promise for enhancing UDA tasks. To address this, we propose a novel method called Cross-Modal Knowledge Distillation (CMKD), leveraging VLP models as teacher models to guide the learning process in the target domain, resulting in state-of-the-art performance. Secondly, current UDA paradigms involve training separate models for each task, leading to significant storage overhead and impractical model deployment as the number of transfer tasks grows. To overcome this challenge, we introduce Residual Sparse Training (RST) exploiting the benefits conferred by VLP s extensive pre-training, a technique that requires minimal adjustment (approximately 0.1%similar to 0.5%) of VLP model parameters to achieve performance comparable to fine-tuning. Combining CMKD and RST, we present a comprehensive solution that effectively leverages VLP models for UDA tasks while reducing storage overhead for model deployment. Furthermore, CMKD can serve as a baseline in conjunction with other methods like FixMatch, enhancing the performance of UDA. Our proposed method outperforms existing techniques on standard benchmarks. Our code will be available at: https://***/Wenlve-Zhou/VLP-UDA.