Lintong Zhang

Christina Kassab, Matias Mattamala, Lintong Zhang et al.

Versatile and adaptive semantic understanding would enable autonomous systems to comprehend and interact with their surroundings. Existing fixed-class models limit the adaptability of indoor mobile and assistive autonomous systems. In this work, we introduce LEXIS, a real-time indoor Simultaneous Localization and Mapping (SLAM) system that harnesses the open-vocabulary nature of Large Language Models (LLMs) to create a unified approach to scene understanding and place recognition. The approach first builds a topological SLAM graph of the environment (using visual-inertial odometry) and embeds Contrastive Language-Image Pretraining (CLIP) features in the graph nodes. We use this representation for flexible room classification and segmentation, serving as a basis for room-centric place recognition. This allows loop closure searches to be directed towards semantically relevant places. Our proposed system is evaluated using both public, simulated data and real-world data, covering office and home environments. It successfully categorizes rooms with varying layouts and dimensions and outperforms the state-of-the-art (SOTA). For place recognition and trajectory estimation tasks we achieve equivalent performance to the SOTA, all also utilizing the same pre-trained model. Lastly, we demonstrate the system's potential for planning.

Lintong Zhang, Yifu Tao, Jiarong Lin et al.

Recent advances in mapping techniques have enabled the creation of highly accurate dense 3D maps during robotic missions, such as point clouds, meshes, or NeRF-based representations. These developments present new opportunities for reusing these maps for localization. However, there remains a lack of a unified approach that can operate seamlessly across different map representations. This paper presents and evaluates a global visual localization system capable of localizing a single camera image across various 3D map representations built using both visual and lidar sensing. Our system generates a database by synthesizing novel views of the scene, creating RGB and depth image pairs. Leveraging the precise 3D geometric map, our method automatically defines rendering poses, reducing the number of database images while preserving retrieval performance. To bridge the domain gap between real query camera images and synthetic database images, our approach utilizes learning-based descriptors and feature detectors. We evaluate the system's performance through extensive real-world experiments conducted in both indoor and outdoor settings, assessing the effectiveness of each map representation and demonstrating its advantages over traditional structure-from-motion (SfM) localization approaches. The results show that all three map representations can achieve consistent localization success rates of 55% and higher across various environments. NeRF synthesized images show superior performance, localizing query images at an average success rate of 72%. Furthermore, we demonstrate an advantage over SfM-based approaches that our synthesized database enables localization in the reverse travel direction which is unseen during the mapping process. Our system, operating in real-time on a mobile laptop equipped with a GPU, achieves a processing rate of 1Hz.

Lintong Zhang, Kang Yin, Seong-Whan Lee

Attribution-based explanation techniques capture key patterns to enhance visual interpretability; however, these patterns often lack the granularity needed for insight in fine-grained tasks, particularly in cases of model misclassification, where explanations may be insufficiently detailed. To address this limitation, we propose a fine-grained counterfactual explanation framework that generates both object-level and part-level interpretability, addressing two fundamental questions: (1) which fine-grained features contribute to model misclassification, and (2) where dominant local features influence counterfactual adjustments. Our approach yields explainable counterfactuals in a non-generative manner by quantifying similarity and weighting component contributions within regions of interest between correctly classified and misclassified samples. Furthermore, we introduce a saliency partition module grounded in Shapley value contributions, isolating features with region-specific relevance. Extensive experiments demonstrate the superiority of our approach in capturing more granular, intuitively meaningful regions, surpassing fine-grained methods.

Yifu Tao, Miguel Ángel Muñoz-Bañón, Lintong Zhang et al.

This paper introduces a large-scale multi-modal dataset captured in and around well-known landmarks in Oxford using a custom-built multi-sensor perception unit as well as a millimetre-accurate map from a Terrestrial LiDAR Scanner (TLS). The perception unit includes three synchronised global shutter colour cameras, an automotive 3D LiDAR scanner, and an inertial sensor - all precisely calibrated. We also establish benchmarks for tasks involving localisation, reconstruction, and novel-view synthesis, which enable the evaluation of Simultaneous Localisation and Mapping (SLAM) methods, Structure-from-Motion (SfM) and Multi-view Stereo (MVS) methods as well as radiance field methods such as Neural Radiance Fields (NeRF) and 3D Gaussian Splatting. To evaluate 3D reconstruction the TLS 3D models are used as ground truth. Localisation ground truth is computed by registering the mobile LiDAR scans to the TLS 3D models. Radiance field methods are evaluated not only with poses sampled from the input trajectory, but also from viewpoints that are from trajectories which are distant from the training poses. Our evaluation demonstrates a key limitation of state-of-the-art radiance field methods: we show that they tend to overfit to the training poses/images and do not generalise well to out-of-sequence poses. They also underperform in 3D reconstruction compared to MVS systems using the same visual inputs. Our dataset and benchmarks are intended to facilitate better integration of radiance field methods and SLAM systems. The raw and processed data, along with software for parsing and evaluation, can be accessed at https://dynamic.robots.ox.ac.uk/datasets/oxford-spires/.

Jianeng Wang, Matias Mattamala, Christina Kassab et al.

Self-balancing exoskeletons are a key enabling technology for individuals with mobility impairments. While the current challenges focus on human-compliant hardware and control, unlocking their use for daily activities requires a scene perception system. In this work, we present Exosense, a vision-centric scene understanding system for self-balancing exoskeletons. We introduce a multi-sensor visual-inertial mapping device as well as a navigation stack for state estimation, terrain mapping and long-term operation. We tested Exosense attached to both a human leg and Wandercraft's Personal Exoskeleton in real-world indoor scenarios. This enabled us to test the system during typical periodic walking gaits, as well as future uses in multi-story environments. We demonstrate that Exosense can achieve an odometry drift of about 4 cm per meter traveled, and construct terrain maps under 1 cm average reconstruction error. It can also work in a visual localization mode in a previously mapped environment, providing a step towards long-term operation of exoskeletons.

Lintong Zhang, Kang Yin, Seong-Whan Lee

In the domain of non-generative visual counterfactual explanations (CE), traditional techniques frequently involve the substitution of sections within a query image with corresponding sections from distractor images. Such methods have historically overlooked the semantic relevance of the replacement regions to the target object, thereby impairing the model's interpretability and hindering the editing workflow. Addressing these challenges, the present study introduces an innovative methodology named as Weighted Semantic Map with Auto-adaptive Candidate Editing Network (WSAE-Net). Characterized by two significant advancements: the determination of an weighted semantic map and the auto-adaptive candidate editing sequence. First, the generation of the weighted semantic map is designed to maximize the reduction of non-semantic feature units that need to be computed, thereby optimizing computational efficiency. Second, the auto-adaptive candidate editing sequences are designed to determine the optimal computational order among the feature units to be processed, thereby ensuring the efficient generation of counterfactuals while maintaining the semantic relevance of the replacement feature units to the target object. Through comprehensive experimentation, our methodology demonstrates superior performance, contributing to a more lucid and in-depth understanding of visual counterfactual explanations.

Lintong Zhang, Marco Camurri, David Wisth et al.

We present a multi-camera LiDAR inertial dataset of 4.5 km walking distance as an expansion of the Newer College Dataset. The global shutter multi-camera device is hardware synchronized with both the IMU and LiDAR, which is more accurate than the original dataset with software synchronization. This dataset also provides six Degrees of Freedom (DoF) ground truth poses at LiDAR frequency (10 Hz). Three data collections are described and an example use case of multi-camera visual-inertial odometry is demonstrated. This expansion dataset contains small and narrow passages, large scale open spaces, as well as vegetated areas, to test localization and mapping systems. Furthermore, some sequences present challenging situations such as abrupt lighting change, textureless surfaces, and aggressive motion. The dataset is available at: https://ori-drs.github. io/newer-college-dataset/

Lintong Zhang, David Wisth, Marco Camurri et al.

We present a multi-camera visual-inertial odometry system based on factor graph optimization which estimates motion by using all cameras simultaneously while retaining a fixed overall feature budget. We focus on motion tracking in challenging environments, such as narrow corridors, dark spaces with aggressive motions, and abrupt lighting changes. These scenarios cause traditional monocular or stereo odometry to fail. While tracking motion with extra cameras should theoretically prevent failures, it leads to additional complexity and computational burden. To overcome these challenges, we introduce two novel methods to improve multi-camera feature tracking. First, instead of tracking features separately in each camera, we track features continuously as they move from one camera to another. This increases accuracy and achieves a more compact factor graph representation. Second, we select a fixed budget of tracked features across the cameras to reduce back-end optimization time. We have found that using a smaller set of informative features can maintain the same tracking accuracy. Our proposed method was extensively tested using a hardware-synchronized device consisting of an IMU and four cameras (a front stereo pair and two lateral) in scenarios including: an underground mine, large open spaces, and building interiors with narrow stairs and corridors. Compared to stereo-only state-of-the-art visual-inertial odometry methods, our approach reduces the drift rate, relative pose error, by up to 80% in translation and 39% in rotation.