Human-Object Interaction Prediction in Videos through Gaze Following

Abstract

Understanding the human-object interactions (HOIs) from a video is essential to fully comprehend a visual scene. This line of research has been addressed by detecting HOIs from images and lately from videos. However, the video-based HOI anticipation task in the third-person view remains understudied. In this paper, we design a framework to detect current HOIs and anticipate future HOIs in videos. We propose to leverage human gaze information since people often fixate on an object before interacting with it. These gaze features together with the scene contexts and the visual appearances of human-object pairs are fused through a spatio-temporal transformer. To evaluate the model in the HOI anticipation task in a multi-person scenario, we propose a set of person-wise multi-label metrics. Our model is trained and validated on the VidHOI dataset, which contains videos capturing daily life and is currently the largest video HOI dataset. Experimental results in the HOI detection task show that our approach improves the baseline by a great margin of 36.3% relatively. Moreover, we conduct an extensive ablation study to demonstrate the effectiveness of our modifications and extensions to the spatio-temporal transformer. Our code is publicly available on https://github.com/nizhf/hoi-prediction-gaze-transformer.

arXiv commentAccepted by CVIU https://doi.org/10.1016/j.cviu.2023.103741

References in corpus (10)

• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Two-Stream Convolutional Networks for Action Recognition in Videos
• Robust Real-World Image Super-Resolution against Adversarial Attacks
• CrowdHuman: A Benchmark for Detecting Human in a Crowd
• Visual Semantic Role Labeling
• LIGHTEN: Learning Interactions with Graph and Hierarchical TEmporal Networks for HOI in videos
• ST-HOI: A Spatial-Temporal Baseline for Human-Object Interaction Detection in Videos
• RED: Reinforced Encoder-Decoder Networks for Action Anticipation
• Video-based Human-Object Interaction Detection from Tubelet Tokens