https://danieledema.github.io/tags/preprint/Recent content in Preprint on Daniele De MartiniHugoen-usMon, 19 Oct 2026 00:00:00 +0000https://danieledema.github.io/publications/20261019-streamrps/Mon, 19 Oct 2026 00:00:00 +0000https://danieledema.github.io/publications/20261019-streamrps/Abstract Robust scene understanding under varying environmental conditions is essential for safe autonomous driving. While cameras and LiDAR degrade in adverse weather and poor lighting, Frequency-Modulated Continuous-Wave (FMCW) spinning radar provides dependable 360° perception at extended ranges with minimal compute cost. However, real-time FMCW radar scene understanding is challenging due to complex signal returns, limited temporal resolution, and scarce annotated datasets. We introduce StreamRPS, a novel framework for streaming FMCW radar panoptic segmentation, comprising an automated label-generation pipeline via LiDAR scene completion and a real-time streaming network for semantic and panoptic segmentation.https://danieledema.github.io/publications/20261016-coredna/Fri, 16 Oct 2026 00:00:00 +0000https://danieledema.github.io/publications/20261016-coredna/Abstract While data collection for model training is relatively easy, curating high-quality datasets from large collections is inherently challenging. Coreset estimation methods address this problem by identifying the best data subsets to train models on. However, existing approaches often require proxy model training, extensive annotations, or task-specific tailoring. Unsupervised alternatives typically exploit geometric properties in foundation model feature spaces, but are limited to pairwise image-to-image comparisons or comparisons against a fixed subset.https://danieledema.github.io/publications/20261004-artemis/Sun, 04 Oct 2026 00:00:00 +0000https://danieledema.github.io/publications/20261004-artemis/Abstract Recent video world models have witnessed the paradigm shift from single-agent to multi-agent involvements, which can reveal more complicated dynamics and cross-agent interaction in the real world. However, existing approaches commonly adopt implicit inter-agent communications via cross attention, which lack explicit geometry constraints and unified 3D state, thereby leading to poor multi-view consistency and struggling with recovering out-of-sight agents. In addition, most of them assume a static background, failing to represent uncontrolled background dynamics.https://danieledema.github.io/publications/20260927-infravla/Sun, 27 Sep 2026 00:00:00 +0000https://danieledema.github.io/publications/20260927-infravla/Abstract Many indoor environments in which robots operate, such as warehouses, offices, and hospitals, already have cameras installed. They observe parts of the building that the robot cannot see from where it stands, yet navigation policies, including recent vision-language-action (VLA) models, do not use them. We propose InfraVLA, an end-to-end method that adapts a pretrained navigation VLA to such static infrastructure views: a closed-circuit television (CCTV) encoder turns each external view into tokens of the input sequence.https://danieledema.github.io/publications/20260921-world_action_model_planning/Mon, 21 Sep 2026 00:00:00 +0000https://danieledema.github.io/publications/20260921-world_action_model_planning/Abstract World action models generate actions together with visual predictions of their consequences. These paired outputs create the potential for planning by sampling multiple actions from one state, comparing their imagined outcomes, and choosing the action with the most promising predicted outcome. However, how to use imagined futures to guide action selection remains unclear. We examine this planning potential empirically. First, we estimate an oracle upper bound on selection by choosing the sampled candidate whose realised outcome is best.https://danieledema.github.io/publications/20260914-curvature_trajectory_evaluation/Mon, 14 Sep 2026 00:00:00 +0000https://danieledema.github.io/publications/20260914-curvature_trajectory_evaluation/Abstract In field robotics, acquiring independent large-scale reference trajectories more accurate than the evaluated estimates remains an open challenge. The domain is widely reliant on Absolute Trajectory Error (ATE) and Relative Pose Error (RPE), computed with automated tools, that rest on assumptions and evaluation parameters rarely made explicit. When unreported, the errors can be misleading and hinder fair comparisons. This paper introduces a trajectory-evaluation protocol for standardized and reliable accuracy assessment in state estimation, localization, and Simultaneous Localization And Mapping (SLAM).https://danieledema.github.io/publications/20260701-spotted/Wed, 01 Jul 2026 00:00:00 +0000https://danieledema.github.io/publications/20260701-spotted/Abstract Animal re-identification (ReID) in camera-trap surveys remains challenging due to low image quality, strong variation in illumination and viewpoint, and highly imbalanced numbers of observations per individual. As a result, current ReID performance is often insufficient for fully automated use, and practical workflows typically depend on expert review of algorithmically proposed candidate matches. Moreover, most existing approaches focus almost exclusively on visual cues and overlook auxiliary information routinely available in field studies, such as image timestamps and camera-trap locations.https://danieledema.github.io/publications/20260624-radar_shadow_geometry/Wed, 24 Jun 2026 00:00:00 +0000https://danieledema.github.io/publications/20260624-radar_shadow_geometry/Abstract Reliable perception in adverse conditions remains challenging for autonomous systems, as cameras and LiDAR degrade in poor lighting or weather. Millimetre-wave FMCW radar is robust to such conditions, but its elevation collapse limits geometric reasoning. We observe that vehicle chassis occlude radar rays and form a distinctive geometric shadow, and its consistency can enable us to infer useful information about objects whose returns intersect this shadow. Motivated by this observation, we propose a method to recover the 3D, in-plane inclination of nearby slender vertical objects from this cue.https://danieledema.github.io/publications/20260603-robotcycle_eye_gaze/Wed, 03 Jun 2026 00:00:00 +0000https://danieledema.github.io/publications/20260603-robotcycle_eye_gaze/Abstract Although much is known about the physical danger of cycling situations, less is understood about the perceived danger of cycling. Furthermore, perception of danger may be filtered at a subconscious level and therefore difficult for one to self-report. To this end, these subconscious perceptions can be revealed through physiological metrics such as eye gaze. This paper explores the perceived safety of cycling in Oxford, United Kingdom and explores the ability of wearable eye tracking glasses to produce insights about the differences in perception under different environments and events.https://danieledema.github.io/publications/20250922-neural_mmgs/Mon, 22 Sep 2025 00:00:00 +0000https://danieledema.github.io/publications/20250922-neural_mmgs/Abstract This paper proposes Neural-MMGS, a novel neural 3DGS framework for multimodal large-scale scene reconstruction that fuses multiple sensing modalities in a per-gaussian compact, learnable embedding. While recent works focusing on large-scale scene reconstruction have incorporated LiDAR data to provide more accurate geometric constraints, we argue that LiDAR’s rich physical properties remain underexplored. Similarly, semantic information has been used for object retrieval, but could provide valuable high-level context for scene reconstruction.https://danieledema.github.io/publications/20241014-nar_icp/Mon, 14 Oct 2024 00:00:00 +0000https://danieledema.github.io/publications/20241014-nar_icp/Abstract This study explores the intersection of neural networks and classical robotics algorithms through the Neural Algorithmic Reasoning (NAR) blueprint, enabling the training of neural networks to reason like classical robotics algorithms by learning to execute them. Algorithms are integral to robotics and safety-critical applications due to their predictable and consistent performance through logical and mathematical principles. In contrast, while neural networks are highly adaptable, handling complex, high-dimensional data and generalising across tasks, they often lack interpretability and transparency in their internal computations.