Learning from Observer Gaze:Zero-Shot Attention Prediction Oriented by Human-Object Interaction Recognition
This addresses a gap in attention prediction for human-object interactions, potentially enhancing human-machine interaction and AI interpretability, though it is incremental as it builds on existing attention research.
The paper tackles the problem of predicting interaction-oriented visual attention, which is unexplored compared to salient instance attention, by introducing a novel gaze dataset IG and a zero-shot task ZeroIA, with their model IA outperforming state-of-the-art methods in both zero-shot and supervised settings.
Most existing attention prediction research focuses on salient instances like humans and objects. However, the more complex interaction-oriented attention, arising from the comprehension of interactions between instances by human observers, remains largely unexplored. This is equally crucial for advancing human-machine interaction and human-centered artificial intelligence. To bridge this gap, we first collect a novel gaze fixation dataset named IG, comprising 530,000 fixation points across 740 diverse interaction categories, capturing visual attention during human observers cognitive processes of interactions. Subsequently, we introduce the zero-shot interaction-oriented attention prediction task ZeroIA, which challenges models to predict visual cues for interactions not encountered during training. Thirdly, we present the Interactive Attention model IA, designed to emulate human observers cognitive processes to tackle the ZeroIA problem. Extensive experiments demonstrate that the proposed IA outperforms other state-of-the-art approaches in both ZeroIA and fully supervised settings. Lastly, we endeavor to apply interaction-oriented attention to the interaction recognition task itself. Further experimental results demonstrate the promising potential to enhance the performance and interpretability of existing state-of-the-art HOI models by incorporating real human attention data from IG and attention labels generated by IA.