Action Scene Graphs for Long-Form Understanding of Egocentric Videos
This addresses the problem of detailed, long-term video analysis for egocentric vision applications, but it is incremental as it builds on existing datasets and annotations.
The paper tackles long-form understanding of egocentric videos by introducing Egocentric Action Scene Graphs (EASGs), a graph-based representation that extends action labels to include objects and temporal evolution, and shows effectiveness in tasks like action anticipation and activity summarization.
We present Egocentric Action Scene Graphs (EASGs), a new representation for long-form understanding of egocentric videos. EASGs extend standard manually-annotated representations of egocentric videos, such as verb-noun action labels, by providing a temporally evolving graph-based description of the actions performed by the camera wearer, including interacted objects, their relationships, and how actions unfold in time. Through a novel annotation procedure, we extend the Ego4D dataset by adding manually labeled Egocentric Action Scene Graphs offering a rich set of annotations designed for long-from egocentric video understanding. We hence define the EASG generation task and provide a baseline approach, establishing preliminary benchmarks. Experiments on two downstream tasks, egocentric action anticipation and egocentric activity summarization, highlight the effectiveness of EASGs for long-form egocentric video understanding. We will release the dataset and the code to replicate experiments and annotations.