arXiv subjects

Bastian Leibe

Publications and source records attributed to Bastian Leibe.

At least 19 recordsLinked to original sources

arXiv2026

Volume Transformer: Revisiting Vanilla Transformers for 3D Scene Understanding

Transformers have become a common foundation across deep learning, yet 3D scene understanding still relies on specialized backbones with strong domain priors. This isolates the field from the broader Transformer ecosystem, limiting the transfer of research advances from other domains and the benefits of increasingly optimized software and hardware stacks. To bridge this gap, we propose the Volume Transformer (Volt), which adapts the vanilla Transformer encoder to 3D scenes with minimal modifications. Specifically, Volt partitions 3D scenes into volumetric patch tokens, processes them with full global self-attention, and injects positional information via 3D rotary positional embeddings (RoPE). Our initial experiments reveal that naively training Volt on standard 3D benchmarks leads to poor generalization, highlighting the limited scale of current 3D supervision. To overcome this, we introduce a data-efficient training recipe based on strong 3D augmentations, regularization, and distillation from a convolutional teacher, making Volt competitive with state-of-the-art methods. We then scale supervision through joint training on multiple datasets and show that Volt benefits more from increased scale than domain-specific 3D backbones, achieving state-of-the-art results on several indoor and outdoor semantic segmentation benchmarks. Finally, as a drop-in backbone in a standard 3D instance segmentation pipeline, Volt also sets a new state of the art, highlighting its potential as a simple, scalable, and general-purpose backbone for 3D scene understanding.

cs.CV↗

arXiv2026

Biased Dreams: Limitations to Epistemic Uncertainty Quantification in Latent Dynamics Models

Model-based reinforcement learning distinguishes between dynamics models operating on proprioceptive states and latent dynamics models typically operating on high-dimensional image observations. Among the latter, Dreamer's Recurrent State Space Model (RSSM) has emerged as a dominant architecture. While ensemble-based epistemic uncertainty has proven effective in proprioceptive dynamics for mitigating model exploitation, guiding exploration, or promoting caution, its behavior in latent dynamics remains largely unexplored. Our experiments reveal that although ensemble disagreement captures local epistemic uncertainty, it does not reliably reflect global compounding model error accumulated over prolonged RSSM latent rollouts. We provide evidence for an attractor behavior that draws rollouts toward well-supported latent regions, where uncertainty diminishes despite increasing discrepancies from the true environment dynamics. This can cause the model to overestimate returns when attractor regions correspond to high-reward behaviors. Our findings reveal a structural limitation of epistemic uncertainty estimation in RSSMs and challenge the assumption that epistemic uncertainty estimation transfers directly from proprioceptive to latent dynamics.

cs.LG↗

arXiv2026

Towards Metric-Agnostic Trajectory Forecasting

Accurate trajectory forecasting of surrounding traffic participants is a core capability for autonomous driving, enabling vehicles to anticipate behavior and plan safe maneuvers. We observe that current state-of-the-art forecasting models on Argoverse 2 and the Waymo Open Motion Dataset tailor their training objectives to the different benchmark metrics. Because these metrics encourage conflicting behavior, we propose a paradigm change for trajectory forecasting: training models with metric-agnostic probabilistic objectives and treating metric optimization as a downstream task applied to the predictive distribution. Concretely, we introduce Trajectory Distribution Evaluation (TraDiE) policies, metric-specific policies that map a predictive distribution to the set of $K$ trajectories and confidences required by trajectory forecasting metrics. We evaluate this framework by introducing DONUT-NLL, which adapts the training objective of the state-of-the-art trajectory forecasting model DONUT to directly optimize the predictive distribution. Using our policies, DONUT-NLL achieves state-of-the-art results on all metrics of the Waymo motion prediction benchmark.

cs.CV↗

arXiv2026

SurGe: Improved Surface Geometry in Point Maps

Recent feedforward 3D reconstruction methods predict point maps and estimate global 3D geometry remarkably well. However, their predictions still exhibit inaccurate local surface geometry, which is clearly visible qualitatively but only weakly reflected in common metrics. To make these errors more explicit in evaluation, we introduce a point map normal metric that evaluates the local surface orientation induced by neighboring 3D predictions. To reduce these errors, we propose two complementary components: a point gradient matching loss that supervises depth-normalized 3D finite differences, and a Neighborhood Attention Decoder (NAD) that progressively upsamples features and uses Neighborhood Attention for local feature mixing. Across eight zero-shot monocular geometry benchmarks, our model, SurGe, achieves the best average rank for global point map AbsRel and consistently improves local point map and point map normal evaluations.

cs.CV↗

arXiv2026

Block-Sparse Global Attention for Efficient Multi-View Geometry Transformers

Efficient and accurate feed-forward multi-view reconstruction has long been an important task in computer vision. Recent transformer-based models like VGGT, $π^3$ and MapAnything have demonstrated remarkable performance with relatively simple architectures. However, their scalability is fundamentally constrained by the quadratic complexity of global attention, which imposes a significant runtime bottleneck when processing large image sets. In this work, we empirically analyze the global attention matrix of these models and observe that the probability mass concentrates on a small subset of patch-patch interactions corresponding to cross-view geometric correspondences. Building on this insight and inspired by recent advances in large language models, we propose a training-free, block-sparse replacement for dense global attention, implemented with highly optimized kernels. Our method accelerates inference by more than $3\times$ while maintaining comparable task performance. Evaluations on a comprehensive suite of multi-view benchmarks demonstrate that our approach seamlessly integrates into existing global attention-based architectures such as VGGT, $π^3$ , and MapAnything, while substantially improving scalability to large image collections.

cs.CV↗

arXiv2026

Query2Uncertainty: Robust Uncertainty Quantification and Calibration for 3D Object Detection under Distribution Shift

Reliable uncertainty estimation for 3D object detection is critical for deploying safe autonomous systems, yet modern detectors remain poorly calibrated, especially under distribution shifts. Although post-hoc calibration methods address this issue and provide improved calibration for in-distribution tests, they fail to adapt in distribution-shifted scenarios. In this work, we address this issue and introduce a density-aware calibration method that couples post-hoc calibrators with the feature density of latent object queries from DETR-style 3D object detectors. These queries form a compact, location and class-aware feature, ideal for density estimation, allowing our approach to adjust model confidences in distribution-shift scenarios. By fitting a density estimator on these query features, our approach jointly recalibrates both classification and bounding box regression uncertainties. On both a multi-view camera and LiDAR-based detector, our approach consistently outperforms standard post-hoc methods in both in-distribution and distribution-shifted scenarios. Code available https://tillbeemelmanns.github.io/query2uncertainty/ .

cs.CV↗

arXiv2026

VidEoMT: Your ViT is Secretly Also a Video Segmentation Model

Existing online video segmentation models typically combine a per-frame segmenter with complex specialized tracking modules. While effective, these modules introduce significant architectural complexity and computational overhead. Recent studies suggest that plain Vision Transformer (ViT) encoders, when scaled with sufficient capacity and large-scale pre-training, can conduct accurate image segmentation without requiring specialized modules. Motivated by this observation, we propose the Video Encoder-only Mask Transformer (VidEoMT), a simple encoder-only video segmentation model that eliminates the need for dedicated tracking modules. To enable temporal modeling in an encoder-only ViT, VidEoMT introduces a lightweight query propagation mechanism that carries information across frames by reusing queries from the previous frame. To balance this with adaptability to new content, it employs a query fusion strategy that combines the propagated queries with a set of temporally-agnostic learned queries. As a result, VidEoMT attains the benefits of a tracker without added complexity, achieving competitive accuracy while being 5x-10x faster, running at up to 160 FPS with a ViT-L backbone. Code: https://www.tue-mps.org/videomt/

cs.CV↗

arXiv2025

DINO in the Room: Leveraging 2D Foundation Models for 3D Segmentation

Vision foundation models (VFMs) trained on large-scale image datasets provide high-quality features that have significantly advanced 2D visual recognition. However, their potential in 3D scene segmentation remains largely untapped, despite the common availability of 2D images alongside 3D point cloud datasets. While significant research has been dedicated to 2D-3D fusion, recent state-of-the-art 3D methods predominantly focus on 3D data, leaving the integration of VFMs into 3D models underexplored. In this work, we challenge this trend by introducing DITR, a generally applicable approach that extracts 2D foundation model features, projects them to 3D, and finally injects them into a 3D point cloud segmentation model. DITR achieves state-of-the-art results on both indoor and outdoor 3D semantic segmentation benchmarks. To enable the use of VFMs even when images are unavailable during inference, we additionally propose to pretrain 3D models by distilling 2D foundation models. By initializing the 3D backbone with knowledge distilled from 2D VFMs, we create a strong basis for downstream 3D segmentation tasks, ultimately boosting performance across various datasets.

cs.CV↗

arXiv2025

Sa2VA-i: Improving Sa2VA Results with Consistent Training and Inference

Sa2VA is a recent model for language-guided dense grounding in images and video that achieves state-of-the-art results on multiple segmentation benchmarks and that has become widely popular. However, we found that Sa2VA does not perform according to its full potential for referring video object segmentation tasks. We identify inconsistencies between training and inference procedures as the key factor holding it back. To mitigate this issue, we propose an improved version of Sa2VA, Sa2VA-i, that rectifies these issues and improves the results. In fact, Sa2VA-i sets a new state of the art for multiple video benchmarks and achieves improvements of up to +11.6 J&F on MeViS, +1.4 on Ref-YT-VOS, +3.3 on Ref-DAVIS and +4.1 on ReVOS using the same Sa2VA checkpoints. With our fixes, the Sa2VA-i-1B model even performs on par with the original Sa2VA-26B model on the MeViS benchmark. We hope that this work will show the importance of seemingly trivial implementation details and that it will provide valuable insights for the referring video segmentation field. We provide the code and updated models at https://github.com/kumuji/sa2va-i

cs.CV↗

arXiv2025

Panoptic-CUDAL: Rural Australia Point Cloud Dataset in Rainy Conditions

Existing autonomous driving datasets are predominantly oriented towards well-structured urban settings and favourable weather conditions, leaving the complexities of rural environments and adverse weather conditions largely unaddressed. Although some datasets encompass variations in weather and lighting, bad weather scenarios do not appear often. Rainfall can significantly impair sensor functionality, introducing noise and reflections in LiDAR and camera data and reducing the system's capabilities for reliable environmental perception and safe navigation. This paper introduces the Panoptic-CUDAL dataset, a novel dataset purpose-built for panoptic segmentation in rural areas subject to rain. By recording high-resolution LiDAR, camera, and pose data, Panoptic-CUDAL offers a diverse, information-rich dataset in a challenging scenario. We present the analysis of the recorded data and provide baseline results for panoptic, semantic segmentation, and 3D occupancy prediction methods on LiDAR point clouds. The dataset can be found here: https://robotics.sydney.edu.au/our-research/intelligent-transportation-systems, https://vision.rwth-aachen.de/panoptic-cudal

cs.CV↗

arXiv2025

LSVOS 2025 Challenge Report: Recent Advances in Complex Video Object Segmentation

This report presents an overview of the 7th Large-scale Video Object Segmentation (LSVOS) Challenge held in conjunction with ICCV 2025. Besides the two traditional tracks of LSVOS that jointly target robustness in realistic video scenarios: Classic VOS (VOS), and Referring VOS (RVOS), the 2025 edition features a newly introduced track, Complex VOS (MOSEv2). Building upon prior insights, MOSEv2 substantially increases difficulty, introducing more challenging but realistic scenarios including denser small objects, frequent disappear/reappear events, severe occlusions, adverse weather and lighting, etc., pushing long-term consistency and generalization beyond curated benchmarks. The challenge retains standard ${J}$, $F$, and ${J\&F}$ metrics for VOS and RVOS, while MOSEv2 adopts ${J\&\dot{F}}$ as the primary ranking metric to better evaluate objects across scales and disappearance cases. We summarize datasets and protocols, highlight top-performing solutions, and distill emerging trends, such as the growing role of LLM/MLLM components and memory-aware propagation, aiming to chart future directions for resilient, language-aware video segmentation in the wild.

cs.CV↗

arXiv2025

DONUT: A Decoder-Only Model for Trajectory Prediction

Predicting the motion of other agents in a scene is highly relevant for autonomous driving, as it allows a self-driving car to anticipate. Inspired by the success of decoder-only models for language modeling, we propose DONUT, a Decoder-Only Network for Unrolling Trajectories. Unlike existing encoder-decoder forecasting models, we encode historical trajectories and predict future trajectories with a single autoregressive model. This allows the model to make iterative predictions in a consistent manner, and ensures that the model is always provided with up-to-date information, thereby enhancing performance. Furthermore, inspired by multi-token prediction for language modeling, we introduce an 'overprediction' strategy that gives the model the auxiliary task of predicting trajectories at longer temporal horizons. This allows the model to better anticipate the future and further improves performance. Through experiments, we demonstrate that our decoder-only approach outperforms the encoder-decoder baseline, and achieves new state-of-the-art results on the Argoverse 2 single-agent motion forecasting benchmark.

cs.CV↗

arXiv2025

How do Foundation Models Compare to Skeleton-Based Approaches for Gesture Recognition in Human-Robot Interaction?

Gestures enable non-verbal human-robot communication, especially in noisy environments like agile production. Traditional deep learning-based gesture recognition relies on task-specific architectures using images, videos, or skeletal pose estimates as input. Meanwhile, Vision Foundation Models (VFMs) and Vision Language Models (VLMs) with their strong generalization abilities offer potential to reduce system complexity by replacing dedicated task-specific modules. This study investigates adapting such models for dynamic, full-body gesture recognition, comparing V-JEPA (a state-of-the-art VFM), Gemini Flash 2.0 (a multimodal VLM), and HD-GCN (a top-performing skeleton-based approach). We introduce NUGGET, a dataset tailored for human-robot communication in intralogistics environments, to evaluate the different gesture recognition approaches. In our experiments, HD-GCN achieves best performance, but V-JEPA comes close with a simple, task-specific classification head - thus paving a possible way towards reducing system complexity, by using it as a shared multi-task model. In contrast, Gemini struggles to differentiate gestures based solely on textual descriptions in the zero-shot setting, highlighting the need of further research on suitable input representations for gestures.

cs.CV↗

arXiv2025

Systematic Comparison of Projection Methods for Monocular 3D Human Pose Estimation on Fisheye Images

Fisheye cameras offer robots the ability to capture human movements across a wider field of view (FOV) than standard pinhole cameras, making them particularly useful for applications in human-robot interaction and automotive contexts. However, accurately detecting human poses in fisheye images is challenging due to the curved distortions inherent to fisheye optics. While various methods for undistorting fisheye images have been proposed, their effectiveness and limitations for poses that cover a wide FOV has not been systematically evaluated in the context of absolute human pose estimation from monocular fisheye images. To address this gap, we evaluate the impact of pinhole, equidistant and double sphere camera models, as well as cylindrical projection methods, on 3D human pose estimation accuracy. We find that in close-up scenarios, pinhole projection is inadequate, and the optimal projection method varies with the FOV covered by the human pose. The usage of advanced fisheye models like the double sphere model significantly enhances 3D human pose estimation accuracy. We propose a heuristic for selecting the appropriate projection model based on the detection bounding box to enhance prediction quality. Additionally, we introduce and evaluate on our novel dataset FISHnCHIPS, which features 3D human skeleton annotations in fisheye images, including images from unconventional angles, such as extreme close-ups, ground-mounted cameras, and wide-FOV poses, available at: https://www.vision.rwth-aachen.de/fishnchips

cs.CV↗

arXiv2025

OpenSplat3D: Open-Vocabulary 3D Instance Segmentation using Gaussian Splatting

3D Gaussian Splatting (3DGS) has emerged as a powerful representation for neural scene reconstruction, offering high-quality novel view synthesis while maintaining computational efficiency. In this paper, we extend the capabilities of 3DGS beyond pure scene representation by introducing an approach for open-vocabulary 3D instance segmentation without requiring manual labeling, termed OpenSplat3D. Our method leverages feature-splatting techniques to associate semantic information with individual Gaussians, enabling fine-grained scene understanding. We incorporate Segment Anything Model instance masks with a contrastive loss formulation as guidance for the instance features to achieve accurate instance-level segmentation. Furthermore, we utilize language embeddings of a vision-language model, allowing for flexible, text-driven instance identification. This combination enables our system to identify and segment arbitrary objects in 3D scenes based on natural language descriptions. We show results on LERF-mask and LERF-OVS as well as the full ScanNet++ validation set, demonstrating the effectiveness of our approach.

cs.CV↗

arXiv2025

How Important are Videos for Training Video LLMs?

Research into Video Large Language Models (LLMs) has progressed rapidly, with numerous models and benchmarks emerging in just a few years. Typically, these models are initialized with a pretrained text-only LLM and finetuned on both image- and video-caption datasets. In this paper, we present findings indicating that Video LLMs are more capable of temporal reasoning after image-only training than one would assume, and that improvements from video-specific training are surprisingly small. Specifically, we show that image-trained versions of two LLMs trained with the recent LongVU algorithm perform significantly above chance level on TVBench, a temporal reasoning benchmark. Additionally, we introduce a simple finetuning scheme involving sequences of annotated images and questions targeting temporal capabilities. This baseline results in temporal reasoning performance close to, and occasionally higher than, what is achieved by video-trained LLMs. This suggests suboptimal utilization of rich temporal features found in real video by current models. Our analysis motivates further research into the mechanisms that allow image-trained LLMs to perform temporal reasoning, as well as into the bottlenecks that render current video training schemes inefficient.

cs.CV↗

arXiv2025

Spotting the Unexpected (STU): A 3D LiDAR Dataset for Anomaly Segmentation in Autonomous Driving

To operate safely, autonomous vehicles (AVs) need to detect and handle unexpected objects or anomalies on the road. While significant research exists for anomaly detection and segmentation in 2D, research progress in 3D is underexplored. Existing datasets lack high-quality multimodal data that are typically found in AVs. This paper presents a novel dataset for anomaly segmentation in driving scenarios. To the best of our knowledge, it is the first publicly available dataset focused on road anomaly segmentation with dense 3D semantic labeling, incorporating both LiDAR and camera data, as well as sequential information to enable anomaly detection across various ranges. This capability is critical for the safe navigation of autonomous vehicles. We adapted and evaluated several baseline models for 3D segmentation, highlighting the challenges of 3D anomaly detection in driving environments. Our dataset and evaluation code will be openly available, facilitating the testing and performance comparison of different approaches.

cs.CV↗

arXiv2025

Acquisition of high-quality images for camera calibration in robotics applications via speech prompts

Accurate intrinsic and extrinsic camera calibration can be an important prerequisite for robotic applications that rely on vision as input. While there is ongoing research on enabling camera calibration using natural images, many systems in practice still rely on using designated calibration targets with e.g. checkerboard patterns or April tag grids. Once calibration images from different perspectives have been acquired and feature descriptors detected, those are typically used in an optimization process to minimize the geometric reprojection error. For this optimization to converge, input images need to be of sufficient quality and particularly sharpness; they should neither contain motion blur nor rolling-shutter artifacts that can arise when the calibration board was not static during image capture. In this work, we present a novel calibration image acquisition technique controlled via voice commands recorded with a clip-on microphone, that can be more robust and user-friendly than e.g. triggering capture with a remote control, or filtering out blurry frames from a video sequence in postprocessing. To achieve this, we use a state-of-the-art speech-to-text transcription model with accurate per-word timestamping to capture trigger words with precise temporal alignment. Our experiments show that the proposed method improves user experience by being fast and efficient, allowing us to successfully calibrate complex multi-camera setups.

cs.RO↗