Recur, Attend or Convolve? On Whether Temporal Modeling Matters for Cross-Domain Robustness in Action Recognition

TL;DR

This paper investigates whether temporal modeling choices, like recurrence, improve cross-domain robustness in action recognition, highlighting the importance of physical inductive biases over purely parameterized models.

Contribution

It introduces the Temporal Shape dataset and modified Diving48 domains to systematically assess the impact of temporal modeling on texture bias and robustness.

Findings

Recurrence may enhance domain shift robustness in action recognition.

Temporal shape cues are crucial for generalization across domains.

Physical inductive biases outperform texture biases in robustness.

Abstract

Most action recognition models today are highly parameterized, and evaluated on datasets with appearance-wise distinct classes. It has also been shown that 2D Convolutional Neural Networks (CNNs) tend to be biased toward texture rather than shape in still image recognition tasks, in contrast to humans. Taken together, this raises suspicion that large video models partly learn spurious spatial texture correlations rather than to track relevant shapes over time to infer generalizable semantics from their movement. A natural way to avoid parameter explosion when learning visual patterns over time is to make use of recurrence. Biological vision consists of abundant recurrent circuitry, and is superior to computer vision in terms of domain shift generalization. In this article, we empirically study whether the choice of low-level temporal modeling has consequences for texture bias and cross-domain robustness. In order to enable a light-weight and systematic assessment of the ability to capture temporal structure, not revealed from single frames, we provide the Temporal Shape (TS) dataset, as well as modified domains of Diving48 allowing for the investigation of spatial texture bias in video models. The combined results of our experiments indicate that sound physical inductive bias such as recurrence in temporal modeling may be advantageous when robustness to domain shift is important for the task.