UnionDet: Union-Level Detector Towards Real-Time Human-Object Interaction Detection
This work addresses the problem of slow inference in HOI detection for computer vision applications, offering a significant speed improvement with competitive accuracy.
The paper tackles the bottleneck in human-object interaction (HOI) detection inference time by proposing UnionDet, a one-stage meta-architecture with a union-level detector that directly captures interaction regions, resulting in a 4x to 14x reduction in interaction prediction time while outperforming state-of-the-art methods on V-COCO and HICO-DET datasets.
Recent advances in deep neural networks have achieved significant progress in detecting individual objects from an image. However, object detection is not sufficient to fully understand a visual scene. Towards a deeper visual understanding, the interactions between objects, especially humans and objects are essential. Most prior works have obtained this information with a bottom-up approach, where the objects are first detected and the interactions are predicted sequentially by pairing the objects. This is a major bottleneck in HOI detection inference time. To tackle this problem, we propose UnionDet, a one-stage meta-architecture for HOI detection powered by a novel union-level detector that eliminates this additional inference stage by directly capturing the region of interaction. Our one-stage detector for human-object interaction shows a significant reduction in interaction prediction time 4x~14x while outperforming state-of-the-art methods on two public datasets: V-COCO and HICO-DET.