English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Recent advances in unsupervised domain adaptation have significantly improvedthe recognition accuracy of CNNs by alleviating the domain shift between(labeled) source and (unlabeled) target data distributions. While the problemof single-target domain adaptation (STDA) for object detection has recentlyreceived much attention, multi-target domain adaptation (MTDA) remains largelyunexplored, despite its practical relevance in several real-world applications,such as multi-camera video surveillance. Compared to the STDA problem that mayinvolve large domain shifts between complex source and target distributions,MTDA faces additional challenges, most notably the computational requirementsand catastrophic forgetting of previously-learned targets, which can depend onthe order of target adaptations. STDA for detection can be applied to MTDA byadapting one model per target, or one common model with a mixture of data fromtarget domains. However, these approaches are either costly or inaccurate. Theonly state-of-art MTDA method specialized for detection learns targetsincrementally, one target at a time, and mitigates the loss of knowledge byusing a duplicated detection model for knowledge distillation, which iscomputationally expensive and does not scale well to many domains. In thispaper, we introduce an efficient approach for incremental learning thatgeneralizes well to multiple target domains. Our MTDA approach is more suitablefor real-world applications since it allows updating the detection modelincrementally, without storing data from previous-learned target domains, norretraining when a new target domain becomes available. Our proposed method,MTDA-DTM, achieved the highest level of detection accuracy compared againststate-of-the-art approaches on several MTDA detection benchmarks and Wildtrack,a benchmark for multi-camera pedestrian detection.

Incremental Multi-Target Domain Adaptation for Object Detection with Efficient Domain Transfer