Niko Sünderhauf
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Recognitions
Niko Sünderhauf has a total of 123 co-authors. Here are the top co-authors by number of shared publications:
- Michael Milford
Queensland University of Technology, Brisbane, Australia
19 shared publications - Feras Dayoub
The University of Adelaide
16 shared publications Peter Protzel
14 shared publicationsBen Upcroft
Oxbotica Ltd., Oxford, UK
10 shared publicationsPeter Corke
Australian Centre for Robotic Vision, Brisbane, Queensland, Australia
7 shared publications- Peter Protzel
Chemnitz University of Technology
6 shared publications
Latest Publications
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Oceanic Engineering
- Vol. 51 (1) , 762-779
Human-in-the-Loop Segmentation of Multispecies Coral Imagery
Marine surveys by robotic underwater and surface vehicles result in substantial quantities of coral reef imagery, however labeling these images is expensive and time-consuming for domain experts. Point label propagation is a technique that uses existing images labeled with sparse points to create augmented ground truth data, which can be used to train a semantic segmentation model. In this work, we show that recent advances in large foundation models facilitate the creation of augmented ground truth masks using only features extracted by the denoised version of the DIstillation of knowledge with NO labels version 2 (DINOv2) foundation model and K-nearest neighbors (KNN), without any pretraining. For images with extremely sparse labels, we use human-in-the-loop principles to enhance annotation efficiency: if there are five point labels per image, our method outperforms the prior state-of-the-art by 19.7% for mean intersection over union (mIoU). When human-in-the-loop labeling is not available, using the denoised DINOv2 features with a KNN still improves on the prior state-of-the-art by 5.8% for mIoU (five grid points). On the semantic segmentation task, we outperform the prior state-of-the-art by 13.5% for mIoU when only five point labels are used for point label propagation. In addition, we perform a comprehensive study into the number and placement of point labels, and make several recommendations for improving the efficiency of labeling images with points.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Oceanic Engineering
- Vol. 51 (1) , 762-779
Human-in-the-Loop Segmentation of Multispecies Coral Imagery
Marine surveys by robotic underwater and surface vehicles result in substantial quantities of coral reef imagery, however labeling these images is expensive and time-consuming for domain experts. Point label propagation is a technique that uses existing images labeled with sparse points to create augmented ground truth data, which can be used to train a semantic segmentation model. In this work, we show that recent advances in large foundation models facilitate the creation of augmented ground truth masks using only features extracted by the denoised version of the DIstillation of knowledge with NO labels version 2 (DINOv2) foundation model and K-nearest neighbors (KNN), without any pretraining. For images with extremely sparse labels, we use human-in-the-loop principles to enhance annotation efficiency: if there are five point labels per image, our method outperforms the prior state-of-the-art by 19.7% for mean intersection over union (mIoU). When human-in-the-loop labeling is not available, using the denoised DINOv2 features with a KNN still improves on the prior state-of-the-art by 5.8% for mIoU (five grid points). On the semantic segmentation task, we outperform the prior state-of-the-art by 13.5% for mIoU when only five point labels are used for point label propagation. In addition, we perform a comprehensive study into the number and placement of point labels, and make several recommendations for improving the efficiency of labeling images with points.
- Open Access
Multi-Modal 3D Scene Graph Updater for Shared and Dynamic Environments
The advent of generalist Large Language Models (LLMs) and Large Vision Models (VLMs) have streamlined the construction of semantically enriched maps that can enable robots to ground high-level reasoning and planning into their representations. One of the most widely used semantic map formats is the 3D Scene Graph, which captures both metric (low-level) and semantic (high-level) information. However, these maps often assume a static world, while real environments, like homes and offices, are dynamic. Even small changes in these spaces can significantly impact task performance. To integrate robots into dynamic environments, they must detect changes and update the scene graph in real-time. This update process is inherently multimodal, requiring input from various sources, such as human agents, the robot's own perception system, time, and its actions. This work proposes a framework that leverages these multimodal inputs to maintain the consistency of scene graphs during real-time operation, presenting promising initial results and outlining a roadmap for future research.
Open AccessMulti-Modal 3D Scene Graph Updater for Shared and Dynamic Environments
The advent of generalist Large Language Models (LLMs) and Large Vision Models (VLMs) have streamlined the construction of semantically enriched maps that can enable robots to ground high-level reasoning and planning into their representations. One of the most widely used semantic map formats is the 3D Scene Graph, which captures both metric (low-level) and semantic (high-level) information. However, these maps often assume a static world, while real environments, like homes and offices, are dynamic. Even small changes in these spaces can significantly impact task performance. To integrate robots into dynamic environments, they must detect changes and update the scene graph in real-time. This update process is inherently multimodal, requiring input from various sources, such as human agents, the robot's own perception system, time, and its actions. This work proposes a framework that leverages these multimodal inputs to maintain the consistency of scene graphs during real-time operation, presenting promising initial results and outlining a roadmap for future research.
- 2 October 2024
- Published by Springer Nature
- p. 1-18
Open-Set Recognition in the Age of Vision-Language Models
- 2 October 2024
- Published by Springer Nature
- p. 1-18
Open-Set Recognition in the Age of Vision-Language Models
- 8 April 2023
- Published by Springer Nature
- p. 177-184
Retrospective
- 8 April 2023
- Published by Springer Nature
- p. 83-103
A Robust Back-End for SLAM
- 8 April 2023
- Published by Springer Nature
- p. 167-170
An Outlook on Robust Optimization for Sensor Fusion and Calibration
- 8 April 2023
- Published by Springer Nature
- p. 167-170
An Outlook on Robust Optimization for Sensor Fusion and Calibration
- 8 April 2023
- Published by Springer Nature
- p. 105-131
Evaluation
- 8 April 2023
- Published by Springer Nature
- p. 75-82
Motivation—When Optimization Fails
- 8 April 2023
- Published by Springer Nature
- p. 45-73
Least Squares Optimization