SADA: Semantic adversarial unsupervised domain adaptation for Temporal Action Localization
It addresses domain shift issues in temporal action localization for real-world applications, representing an incremental advance with novel benchmarks and a method for finer-grained adaptation.
The paper tackled the problem of generalizing temporal action localization to unseen domains by introducing SADA, an unsupervised domain adaptation approach for sparse action detection, achieving a performance boost of up to 6.14% mAP compared to state-of-the-art methods.
Temporal Action Localization (TAL) is a complex task that poses relevant challenges, particularly when attempting to generalize on new -- unseen -- domains in real-world applications. These scenarios, despite realistic, are often neglected in the literature, exposing these solutions to important performance degradation. In this work, we tackle this issue by introducing, for the first time, an approach for Unsupervised Domain Adaptation (UDA) in sparse TAL, which we refer to as Semantic Adversarial unsupervised Domain Adaptation (SADA). Our contributions are threefold: (1) we pioneer the development of a domain adaptation model that operates on realistic sparse action detection benchmarks; (2) we tackle the limitations of global-distribution alignment techniques by introducing a novel adversarial loss that is sensitive to local class distributions, ensuring finer-grained adaptation; and (3) we present a novel set of benchmarks based on EpicKitchens100 and CharadesEgo, that evaluate multiple domain shifts in a comprehensive manner. Our experiments indicate that SADA improves the adaptation across domains when compared to fully supervised state-of-the-art and alternative UDA methods, attaining a performance boost of up to 6.14% mAP.