Hue Histograms to Spatiotemporal Local Features for Action Recognition

TL;DR

This paper introduces HueSTIP, a color-aware extension of the STIP descriptor, which incorporates local color information to improve action recognition in videos, showing potential performance enhancements over the original method.

Contribution

The paper presents a novel color-aware spatiotemporal feature descriptor, HueSTIP, extending STIP to include local hue information for better action recognition.

Findings

HueSTIP outperforms original STIP in action recognition tasks.

Color information significantly improves spatiotemporal feature descriptors.

HueSTIP demonstrates the importance of local color cues in video analysis.

Abstract

Despite the recent developments in spatiotemporal local features for action recognition in video sequences, local color information has so far been ignored. However, color has been proved an important element to the success of automated recognition of objects and scenes. In this paper we extend the space-time interest point descriptor STIP to take into account the color information on the features' neighborhood. We compare the performance of our color-aware version of STIP (which we have called HueSTIP) with the original one.