What to look at and where: Semantic and Spatial Refined Transformer for detecting human-object interactions
This work addresses the problem of accurately detecting interactions between humans and objects in images for computer vision applications, representing an incremental improvement over existing Transformer-based methods.
The authors tackled human-object interaction detection by proposing a semantic and spatial refined transformer (SSRT) that introduces modules to select relevant object-action pairs and refine query representations, achieving state-of-the-art results on V-COCO and HICO-DET benchmarks.
We propose a novel one-stage Transformer-based semantic and spatial refined transformer (SSRT) to solve the Human-Object Interaction detection task, which requires to localize humans and objects, and predicts their interactions. Differently from previous Transformer-based HOI approaches, which mostly focus at improving the design of the decoder outputs for the final detection, SSRT introduces two new modules to help select the most relevant object-action pairs within an image and refine the queries' representation using rich semantic and spatial features. These enhancements lead to state-of-the-art results on the two most popular HOI benchmarks: V-COCO and HICO-DET.