Learn from the Learnt: Source-Free Active Domain Adaptation via Contrastive Sampling and Visual Persistence

TL;DR

This paper introduces a novel source-free active domain adaptation method called LFTL that leverages learned knowledge and persistent challenging samples to improve target domain adaptation without source data, achieving state-of-the-art results.

Contribution

The paper proposes a new paradigm for source-free active domain adaptation that uses contrastive active sampling and visual persistence-guided adaptation, with no extra overhead.

Findings

Achieves state-of-the-art performance on benchmarks.

Demonstrates superior computational efficiency.

Shows continuous improvement with increased annotation budget.

Abstract

Domain Adaptation (DA) facilitates knowledge transfer from a source domain to a related target domain. This paper investigates a practical DA paradigm, namely Source data-Free Active Domain Adaptation (SFADA), where source data becomes inaccessible during adaptation, and a minimum amount of annotation budget is available in the target domain. Without referencing the source data, new challenges emerge in identifying the most informative target samples for labeling, establishing cross-domain alignment during adaptation, and ensuring continuous performance improvements through the iterative query-and-adaptation process. In response, we present learn from the learnt (LFTL), a novel paradigm for SFADA to leverage the learnt knowledge from the source pretrained model and actively iterated models without extra overhead. We propose Contrastive Active Sampling to learn from the hypotheses of the preceding model, thereby querying target samples that are both informative to the current model and persistently challenging throughout active learning. During adaptation, we learn from features of actively selected anchors obtained from previous intermediate models, so that the Visual Persistence-guided Adaptation can facilitate feature distribution alignment and active sample exploitation. Extensive experiments on three widely-used benchmarks show that our LFTL achieves state-of-the-art performance, superior computational efficiency and continuous improvements as the annotation budget increases. Our code is available at https://github.com/lyumengyao/lftl.