HOTR: End-to-End Human-Object Interaction Detection with Transformers
This addresses the bottleneck of time-consuming post-processing in HOI detection for computer vision applications, representing a novel method rather than an incremental improvement.
The paper tackles the problem of Human-Object Interaction (HOI) detection by introducing HOTR, a framework that directly predicts interaction triplets using transformers, achieving state-of-the-art performance on two benchmarks with inference under 1 ms after object detection.
Human-Object Interaction (HOI) detection is a task of identifying "a set of interactions" in an image, which involves the i) localization of the subject (i.e., humans) and target (i.e., objects) of interaction, and ii) the classification of the interaction labels. Most existing methods have indirectly addressed this task by detecting human and object instances and individually inferring every pair of the detected instances. In this paper, we present a novel framework, referred to by HOTR, which directly predicts a set of <human, object, interaction> triplets from an image based on a transformer encoder-decoder architecture. Through the set prediction, our method effectively exploits the inherent semantic relationships in an image and does not require time-consuming post-processing which is the main bottleneck of existing methods. Our proposed algorithm achieves the state-of-the-art performance in two HOI detection benchmarks with an inference time under 1 ms after object detection.