Next-Active-Object prediction from Egocentric Videos
This addresses the challenge of anticipating user intentions for intelligent wearable systems, though it is incremental as it builds on existing First Person Vision paradigms.
The paper tackles the problem of predicting which object a user will interact with next from egocentric videos, by training a classifier to recognize trajectories leading to object activation and forecasting next-active-objects using a temporal sliding window, achieving favorable results compared to baselines on the ADL dataset with 10 hours of video from 20 subjects.
Although First Person Vision systems can sense the environment from the user's perspective, they are generally unable to predict his intentions and goals. Since human activities can be decomposed in terms of atomic actions and interactions with objects, intelligent wearable systems would benefit from the ability to anticipate user-object interactions. Even if this task is not trivial, the First Person Vision paradigm can provide important cues to address this challenge. We propose to exploit the dynamics of the scene to recognize next-active-objects before an object interaction begins. We train a classifier to discriminate trajectories leading to an object activation from all others and forecast next-active-objects by analyzing fixed-length trajectory segments within a temporal sliding window. The proposed method compares favorably with respect to several baselines on the Activity of Daily Living (ADL) egocentric dataset comprising 10 hours of videos acquired by 20 subjects while performing unconstrained interactions with several objects.