# Timeception for Complex Action Recognition

## TL;DR

This paper introduces Timeception, a novel multi-scale temporal convolution approach that effectively captures long-range temporal dependencies in complex human activities within videos, significantly improving recognition accuracy.

## Contribution

The paper proposes Timeception layers that reduce 3D convolution complexity and model minute-long temporal patterns, advancing long-range temporal modeling in activity recognition.

## Key findings

- Timeception outperforms existing methods on Charades, Breakfast Actions, and MultiTHUMOS datasets.
- It learns long-range temporal dependencies effectively.
- The approach captures temporal extents up to eight times longer than previous models.

## Abstract

This paper focuses on the temporal aspect for recognizing human activities in videos; an important visual cue that has long been undervalued. We revisit the conventional definition of activity and restrict it to Complex Action: a set of one-actions with a weak temporal pattern that serves a specific purpose. Related works use spatiotemporal 3D convolutions with fixed kernel size, too rigid to capture the varieties in temporal extents of complex actions, and too short for long-range temporal modeling. In contrast, we use multi-scale temporal convolutions, and we reduce the complexity of 3D convolutions. The outcome is Timeception convolution layers, which reasons about minute-long temporal patterns, a factor of 8 longer than best related works. As a result, Timeception achieves impressive accuracy in recognizing the human activities of Charades, Breakfast Actions, and MultiTHUMOS. Further, we demonstrate that Timeception learns long-range temporal dependencies and tolerate temporal extents of complex actions.