Leveraging Next-Active Objects for Context-Aware Anticipation in Egocentric Videos

TL;DR

This paper introduces NAOGAT, a transformer-based model that leverages object dynamics and scene context to predict future human-object interactions and actions in egocentric videos, outperforming existing methods.

Contribution

The paper presents a novel multi-modal transformer architecture that jointly models object and scene context for improved anticipation of future actions and objects in egocentric videos.

Findings

NAOGAT outperforms existing methods on Ego4D and EpicKitchens-100 datasets.

The model effectively exploits object motion dynamics for better anticipation.

It achieves higher accuracy in predicting time to contact and next-active-object localization.

Abstract

Objects are crucial for understanding human-object interactions. By identifying the relevant objects, one can also predict potential future interactions or actions that may occur with these objects. In this paper, we study the problem of Short-Term Object interaction anticipation (STA) and propose NAOGAT (Next-Active-Object Guided Anticipation Transformer), a multi-modal end-to-end transformer network, that attends to objects in observed frames in order to anticipate the next-active-object (NAO) and, eventually, to guide the model to predict context-aware future actions. The task is challenging since it requires anticipating future action along with the object with which the action occurs and the time after which the interaction will begin, a.k.a. the time to contact (TTC). Compared to existing video modeling architectures for action anticipation, NAOGAT captures the relationship between objects and the global scene context in order to predict detections for the next active object and anticipate relevant future actions given these detections, leveraging the objects' dynamics to improve accuracy. One of the key strengths of our approach, in fact, is its ability to exploit the motion dynamics of objects within a given clip, which is often ignored by other models, and separately decoding the object-centric and motion-centric information. Through our experiments, we show that our model outperforms existing methods on two separate datasets, Ego4D and EpicKitchens-100 ("Unseen Set"), as measured by several additional metrics, such as time to contact, and next-active-object localization. The code will be available upon acceptance.