The Instantaneous Accuracy: a Novel Metric for the Problem of Online Human Behaviour Recognition in Untrimmed Videos

The problem of Online Human Behavior Recognition in untrimmed videos, aka On-line Action Detection (OAI)), needs to he revisited. Unlike traditional off-line action detection approaches, where the evaluation metrics are clear and well established, in the OAI) setting we find few works and no consensus on evaluation protocols to be used. In this paper we introduce a novel on-line metric, the Instantaneous Accuracy (IA), that exhibits an on-line nature, solving most of the limitations of the previous (off-line) metrics. We conduct a thorough experimental evaluation on the TI/Series dataset, comparing the performance of various baseline methods with the state of the art. Our results confirm the problems of the previous evaluation protocols, and suggest that an IA-based protocol is more adequate to the on-line scenario for human behavior understanding.