H2O: Two Hands Manipulating Objects for First Person Interaction Recognition

TL;DR

This paper introduces H2O, a comprehensive egocentric dataset with 3D annotations of two hands and objects, and proposes a method to recognize first-person interactions by estimating hand and object poses from RGB images.

Contribution

The paper presents the first detailed egocentric 3D interaction dataset with synchronized multi-view RGB-D data and a novel pose-based interaction recognition method using graph convolutional networks.

Findings

Achieved state-of-the-art accuracy in first-person interaction recognition.

Established a strong baseline for joint hand-object pose estimation.

Provided a new benchmark dataset for egocentric 3D interaction analysis.

Abstract

We present a comprehensive framework for egocentric interaction recognition using markerless 3D annotations of two hands manipulating objects. To this end, we propose a method to create a unified dataset for egocentric 3D interaction recognition. Our method produces annotations of the 3D pose of two hands and the 6D pose of the manipulated objects, along with their interaction labels for each frame. Our dataset, called H2O (2 Hands and Objects), provides synchronized multi-view RGB-D images, interaction labels, object classes, ground-truth 3D poses for left & right hands, 6D object poses, ground-truth camera poses, object meshes and scene point clouds. To the best of our knowledge, this is the first benchmark that enables the study of first-person actions with the use of the pose of both left and right hands manipulating objects and presents an unprecedented level of detail for egocentric 3D interaction recognition. We further propose the method to predict interaction classes by estimating the 3D pose of two hands and the 6D pose of the manipulated objects, jointly from RGB images. Our method models both inter- and intra-dependencies between both hands and objects by learning the topology of a graph convolutional network that predicts interactions. We show that our method facilitated by this dataset establishes a strong baseline for joint hand-object pose estimation and achieves state-of-the-art accuracy for first person interaction recognition.