HD-EPIC: A Highly-Detailed Egocentric Video Dataset

TL;DR

HD-EPIC is a comprehensive egocentric video dataset with detailed 3D annotations, enabling advanced research in recipe understanding, action recognition, and scene perception in real-world kitchen environments.

Contribution

The paper introduces HD-EPIC, a novel in-the-wild egocentric dataset with detailed 3D annotations, including recipes, actions, and scene information, supporting diverse vision tasks.

Findings

Powerful annotations enable challenging VQA benchmarks.

Current VLMs struggle with detailed egocentric data.

HD-EPIC supports multiple vision tasks like action and sound recognition.

Abstract

We present a validation dataset of newly-collected kitchen-based egocentric videos, manually annotated with highly detailed and interconnected ground-truth labels covering: recipe steps, fine-grained actions, ingredients with nutritional values, moving objects, and audio annotations. Importantly, all annotations are grounded in 3D through digital twinning of the scene, fixtures, object locations, and primed with gaze. Footage is collected from unscripted recordings in diverse home environments, making HDEPIC the first dataset collected in-the-wild but with detailed annotations matching those in controlled lab environments. We show the potential of our highly-detailed annotations through a challenging VQA benchmark of 26K questions assessing the capability to recognise recipes, ingredients, nutrition, fine-grained actions, 3D perception, object motion, and gaze direction. The powerful long-context Gemini Pro only achieves 38.5% on this benchmark, showcasing its difficulty and highlighting shortcomings in current VLMs. We additionally assess action recognition, sound recognition, and long-term video-object segmentation on HD-EPIC. HD-EPIC is 41 hours of video in 9 kitchens with digital twins of 413 kitchen fixtures, capturing 69 recipes, 59K fine-grained actions, 51K audio events, 20K object movements and 37K object masks lifted to 3D. On average, we have 263 annotations per minute of our unscripted videos.