Continually adapt model to new domains without catastrophic forget using visual prompts with CTAP
Continually adapt model to new domains without catastrophic forget using visual prompts with CTAP
Decorate the Newcomers: Visual Domain Prompt for Continual Test Time Adaptation
arXiv paper abstract https://arxiv.org/abs/2212.04145
arXiv PDF paper https://arxiv.org/pdf/2212.04145.pdf
Continual Test-Time Adaptation (CTTA) aims to adapt the source model to continually changing unlabeled target domains without access to the source data.
Existing methods … such as predicting pseudo labels … are noisy and unreliable, these methods suffer from catastrophic forgetting and error accumulation when dealing with dynamic data distributions.
Motivated by the prompt learning in NLP … During testing, the changing target datasets can be adapted to the source model by reformulating the input data with the learned visual prompts.
… devise two types of prompts, i.e., domains-specific prompts and domains-agnostic prompts, to extract current domain knowledge and maintain the domain-shared knowledge in the continual adaptation.
Furthermore, … design a homeostasis-based prompt adaptation strategy to suppress domain-sensitive parameters in domain-invariant prompts to learn domain-shared knowledge more effectively.
… proposed method achieves significant performance gains over state-of-the-art methods …
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
LinkedIn https://www.linkedin.com/in/morris-lee-47877b7b
