Efficient Multi-Task RGB-D Scene Analysis for Indoor Environments

TL;DR

This paper introduces EMSANet, an efficient multi-task neural network that performs semantic segmentation, instance segmentation, orientation estimation, and scene classification simultaneously on RGB-D data, enabling real-time indoor scene analysis on mobile devices.

Contribution

The paper presents EMSANet, the first multi-task model for comprehensive indoor scene analysis on RGB-D data that operates in real time on mobile platforms.

Findings

All tasks are achieved with a single neural network in real time.

Multi-task learning improves performance across tasks.

Extended annotations enable comprehensive evaluation.

Abstract

Semantic scene understanding is essential for mobile agents acting in various environments. Although semantic segmentation already provides a lot of information, details about individual objects as well as the general scene are missing but required for many real-world applications. However, solving multiple tasks separately is expensive and cannot be accomplished in real time given limited computing and battery capabilities on a mobile platform. In this paper, we propose an efficient multi-task approach for RGB-D scene analysis~(EMSANet) that simultaneously performs semantic and instance segmentation~(panoptic segmentation), instance orientation estimation, and scene classification. We show that all tasks can be accomplished using a single neural network in real time on a mobile platform without diminishing performance - by contrast, the individual tasks are able to benefit from each other. In order to evaluate our multi-task approach, we extend the annotations of the common RGB-D indoor datasets NYUv2 and SUNRGB-D for instance segmentation and orientation estimation. To the best of our knowledge, we are the first to provide results in such a comprehensive multi-task setting for indoor scene analysis on NYUv2 and SUNRGB-D.