Neural radiance fields are an emerging 3D scene representation and recently even been extended to learn features for scene understanding by distilling open-vocabulary features from vision-language models. However, current method primarily focus on object-centric representations, supporting object segmentation or detection, while understanding semantic relationships between objects remains largely unexplored. To address this gap, we propose RelationField, the first method to extract inter-object relationships directly from neural radiance fields. RelationField represents relationships between objects as pairs of rays within a neural radiance field, effectively extending its formulation to include implicit relationship queries. To teach RelationField complex, open-vocabulary relationships, relationship knowledge is distilled from multi-modal LLMs. To evaluate RelationField, we solve open-vocabulary 3D scene graph generation tasks and relationship-guided instance segmentation, achieving state-of-the-art performance in both tasks. See the project website at relationfield.github.io.

3D scene graphs are an emerging 3D scene representation, that models both the objects present in the scene as well as their relationships. However, learning 3D scene graphs is a challenging task because it requires not only object labels but also relationship annotations, which are very scarce in datasets. While it is widely accepted that pre-training is an effective approach to improve model performance in low data regimes, in this paper, we find that existing pre-training methods are ill-suited for 3D scene graphs. To solve this issue, we present the first language-based pre-training approach for 3D scene graphs, whereby we exploit the strong relationship between scene graphs and language. To this end, we leverage the language encoder of CLIP, a popular vision-language model, to distill its knowledge into our graph-based network. We formulate a contrastive pre-training, which aligns text embeddings of relationships (subject-predicate-object triplets) and predicted 3D graph features. Our method achieves state-of-the-art results on the main semantic 3D scene graph benchmark by showing improved effectiveness over pre-training baselines and outperforming all the existing fully supervised scene graph prediction methods by a significant margin. Furthermore, since our scene graph features are language-aligned, it allows us to query the language space of the features in a zero-shot manner. In this paper, we show an example of utilizing this property of the features to predict the room type of a scene without further training.

Current approaches for 3D scene graph prediction rely on labeled datasets to train models for a fixed set of known object classes and relationship categories. We present Open3DSG, an alternative approach to learn 3D scene graph prediction in an open world without requiring labeled scene graph data. We co-embed the features from a 3D scene graph prediction backbone with the feature space of pow-erful open world 2D vision language foundation models. This enables us to predict 3D scene graphs from 3D point clouds in a zero-shot manner by querying object classes from an open vocabulary and predicting the inter-object relationships from a grounded LLM with scene graph features and queried object classes as context. Open3DSG is the first 3D point cloud method to predict not only explicit open-vocabulary object classes, but also open-set relation-ships that are not limited to a predefined label set, making it possible to express rare as well as specific objects and relationships in the predicted 3D scene graph. Our exper-iments show that Open3DSG is effective at predicting arbitrary object classes as well as their complex inter-object relationships describing spatial, supportive, semantic and comparative relationships.

In the field of 3D scene understanding, 3D scene graphs have emerged as a new scene representation that combines geometric and semantic information about objects and their relationships. However, learning semantic 3D scene graphs in a fully supervised manner is inherently difficult as it requires not only object-level annotations but also relationship labels. While pre-training approaches have helped to boost the performance of many methods in various fields, pre-training for 3D scene graph prediction has received little attention. Furthermore, we find in this paper that classical contrastive point cloud-based pre-training approaches are ineffective for 3D scene graph learning. To this end, we present SGRec3D, a novel self-supervised pre-training method for 3D scene graph prediction. We propose to reconstruct the 3D input scene from a graph bottleneck as a pretext task. Pre-training SGRec3D does not require object relationship labels, making it possible to exploit large-scale 3D scene understanding datasets, which were off-limits for 3D scene graph learning before. Our experiments demonstrate that in contrast to recent point cloud-based pre-training approaches, our proposed pre-training improves the 3D scene graph prediction considerably, which results in SOTA performance, outperforming other 3D scene graph models by +10% on object prediction and +4% on relationship prediction. Additionally, we show that only using a small subset of 10% labeled data during fine-tuning is sufficient to outperform the same model without pre-training.

Simultaneous Localization and Mapping (SLAM) is considered a mature research field with numerous applications and publicly available open-source systems. Despite this maturity, existing SLAM systems often rely on ad-hoc implementations or are tailored to predefined sensor setups. In this work, we tackle these issues, proposing a novel unified SLAM architecture specifically designed to standardize the SLAM problem and to address heterogeneous sensor configurations. Thanks to its modularity and design patterns, the presented framework is easy to extend, maximizes code reuse and improves computational efficiency. We show in our experiments with a variety of typical sensor configurations that these advantages come without compromising state-of-the-art SLAM performance. The result demonstrates the architecture's relevance for facilitating further research in (multi-sensor) SLAM and its transfer into practical applications.

A Simultaneous Localization and Mapping (SLAM) system is a complex program consisting of several interconnected components with different functionalities such as optimization, tracking or loop detection. Whereas the literature addresses in detail how enhancing the algorithmic aspects of the individual components improves SLAM performance, the modal aspects, such as when to localize, relocalize or close a loop, are usually left aside. In this paper, we address the modal aspects of a SLAM system and show that the design of the modal controller has a strong impact on SLAM performance in particular in terms of robustness against unforeseen events such as sensor failures, perceptual aliasing or kidnapping. We preset a novel taxonomy for the components of a modern SLAM system, investigate their interplay and propose a highly modular architecture of a generic SLAM system using the Unified Modeling LanguageTM(UML) state machine formalism. The result, called SLAM state machine, is compared to the modal controller of several state-of-the-art SLAM systems and evaluated in two experiments. We demonstrate that our state machine handles unforeseen events much more robustly than the state-of-the-art systems.

In this paper we present ProSLAM, a lightweight open-source stereo visual SLAM system designed with simplicity in mind. This work stems from the experience gathered by the authors while teaching SLAM and aims at providing a highly modular system that can be easily implemented and understood. Rather than focusing on the well known mathematical aspects of stereo visual SLAM, we highlight the data structures and the algorithmic aspects required to realize such a system. We implemented ProSLAM using the C++ programming language in combination with a minimal set of standard libraries. The results of a thorough validation performed on several standard benchmark datasets show that ProSLAM achieves precision comparable to state-of-the-art approaches, while requiring substantially less computation.