# High-quality Instance-aware Semantic 3D Map Using RGB-D Camera

Source: [High-quality Instance-aware Semantic 3D Map Using RGB-D Camera](<https://www.sota2.com/research/paper/68fa66ad1e209131cea11a36>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T21:55:21.281Z

Authors: Dinh-Cuong Hoang, Todor Stoyanov, Achim J. Lilienthal

Published: 2019-03-26T10:42:09.000Z

## Abstract

We present a mapping system capable of constructing detailed instance-level semantic models of room-sized indoor environments by means of an RGB-D camera. In this work, we integrate deep-learning-based instance segmentation and classification into a state of the art RGB-D SLAM system. We leverage the pipeline of ElasticFusion \[1\] as a backbone and propose modifications of the registration cost function. The proposed objective function features a tunable weight for the appearance channel, which can be learned from data. The resulting system is capable of producing accurate semantic maps of room-sized environments, as well as reconstructing highly detailed object-level models. The developed method has been verified through experimental validation on the TUMRGB-D SLAM benchmark and the YCB video dataset. Our results confirmed that the proposed system performs favorably in terms of trajectory estimation, surface reconstruction, and segmentation quality in comparison to other state-of-the-art systems.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1903.10782>)

[PDF](<https://arxiv.org/pdf/1903.10782.pdf>)
