Delving Deep into One-Shot Skeleton-based Action Recognition with Diverse Occlusions

TL;DR

This paper introduces a new benchmark for one-shot skeleton-based action recognition under realistic occlusions, evaluates existing models, and proposes a transformer-based model that improves robustness against occlusions.

Contribution

It formalizes the first benchmark for occluded skeleton-based one-shot action recognition and introduces Trans4SOAR, a transformer model that enhances recognition accuracy under occlusions.

Findings

Trans4SOAR outperforms existing models on all datasets.

Recognition accuracy declines with occlusions, but less so with Trans4SOAR.

State-of-the-art performance achieved on NTU-120 without occlusion.

Abstract

Occlusions are universal disruptions constantly present in the real world. Especially for sparse representations, such as human skeletons, a few occluded points might destroy the geometrical and temporal continuity critically affecting the results. Yet, the research of data-scarce recognition from skeleton sequences, such as one-shot action recognition, does not explicitly consider occlusions despite their everyday pervasiveness. In this work, we explicitly tackle body occlusions for Skeleton-based One-shot Action Recognition (SOAR). We mainly consider two occlusion variants: 1) random occlusions and 2) more realistic occlusions caused by diverse everyday objects, which we generate by projecting the existing IKEA 3D furniture models into the camera coordinate system of the 3D skeletons with different geometric parameters. We leverage the proposed pipeline to blend out portions of skeleton sequences of the three popular action recognition datasets and formalize the first benchmark for SOAR from partially occluded body poses. Another key property of our benchmark are the more realistic occlusions generated by everyday objects, as even in standard recognition from 3D skeletons, only randomly missing joints were considered. We re-evaluate existing state-of-the-art frameworks for SOAR in the light of this new task and further introduce Trans4SOAR - a new transformer-based model which leverages three data streams and mixed attention fusion mechanism to alleviate the adverse effects caused by occlusions. While our experiments demonstrate a clear decline in accuracy with missing skeleton portions, this effect is smaller with Trans4SOAR, which outperforms other architectures on all datasets. Although we specifically focus on occlusions, Trans4SOAR additionally yields state-of-the-art in the standard SOAR without occlusion, surpassing the best published approach by 2.85% on NTU-120.