Temporal Pyramid Network for Action Recognition

TL;DR

This paper introduces a flexible feature-level Temporal Pyramid Network (TPN) that enhances action recognition by capturing various visual tempos, achieving consistent improvements across multiple datasets and backbone architectures.

Contribution

The paper proposes a novel, plug-and-play feature-level TPN that models visual tempo hierarchies within backbone networks for improved action recognition.

Findings

2% accuracy gain on Kinetics-400 validation set with TPN

Most improvements occur on actions with high tempo variance

TPN can be integrated into 2D or 3D networks easily

Abstract

Visual tempo characterizes the dynamics and the temporal scale of an action. Modeling such visual tempos of different actions facilitates their recognition. Previous works often capture the visual tempo through sampling raw videos at multiple rates and constructing an input-level frame pyramid, which usually requires a costly multi-branch network to handle. In this work we propose a generic Temporal Pyramid Network (TPN) at the feature-level, which can be flexibly integrated into 2D or 3D backbone networks in a plug-and-play manner. Two essential components of TPN, the source of features and the fusion of features, form a feature hierarchy for the backbone so that it can capture action instances at various tempos. TPN also shows consistent improvements over other challenging baselines on several action recognition datasets. Specifically, when equipped with TPN, the 3D ResNet-50 with dense sampling obtains a 2% gain on the validation set of Kinetics-400. A further analysis also reveals that TPN gains most of its improvements on action classes that have large variances in their visual tempos, validating the effectiveness of TPN.