#01Nov 16, 2026
cs.CV
FM-Recon: Exploring Foundation Model Features for Accelerated Cardiac Cine MRI Reconstruction
Donghang Lyu, Marius Staring, Ruochen Gao and 3 more
Foundation models have demonstrated strong feature extraction capabilities across various medical imaging tasks, yet their potential for MRI reconstruction remains largely unexplored, particularly within unrolled reconstruction networks and under artifact-corrupted inputs. In this work, we explore whether and how features from pretrained foundation models can benefit accelerated cardiac cine MRI reconstruction. To this end, we propose FM-Recon, a framework that refines and integrates extracted features from frozen image encoder of foundation models into an unrolled reconstruction network. Through extensive experiments on the CMRxRecon2023 dataset, we systematically evaluate foundation model features across different reconstruction settings. Our results show that foundation models can extract informative representations from artifact-corrupted inputs and improve reconstruction performance, particularly at moderate acceleration factors. These findings demonstrate the potential of foundation model features for accelerated MRI reconstruction and provide insights into their effective integration into unrolled reconstruction networks.
#02Nov 16, 2026
physics.chem-ph
Rotation-Invariant Machine Learning for Highly Symmetric Molecular Configurations
Logan Bolton, Tomas Karella, Nicholas Lubbers and 5 more
Atomistic machine learning can approximate expensive quantum mechanical calculations at a fraction of their computational cost, but accuracy depends strongly on how the three-dimensional atomic environments are represented. Invariant representations enable fast computation but can fail to distinguish geometrically distinct configurations, especially in highly symmetric environments. We introduce a lightweight benchmark to evaluate expressivity on $n$-fold symmetric configurations, and use it to expose degeneracies in both invariant and equivariant architectures. Guided by these results, we construct a new invariant basis for HIP-HOP which resolves most of the identified failures. On a methane simulation dataset, the extended basis achieves the lowest error among the HIP-HOP variants and has similar accuracy to EquiformerV3 while running $3.2\times$ faster at inference. These results show that carefully constructed invariant representations can mitigate key expressivity limitations while balancing accuracy and computational efficiency.
#03Nov 16, 2026
cs.AI
CivilBench: Benchmarking Vision-Language Models for Multimodal Information Integration in Civil Engineering
Dan Yuan, Yue Wen, Zhili He and 3 more
Artificial intelligence (AI), particularly large language models (LLMs) and vision-language models (VLMs), is increasingly applied in civil and structural engineering for disaster assessment, structural analysis, and engineering decision support. However, the extent to which current VLMs possess the reasoning capabilities required for reliable engineering judgments remains unclear. In this work, we introduce CivilBench, a comprehensive benchmark for evaluating VLMs in civil engineering and disaster-related scenarios. CivilBench covers three major task families: Sequential Reasoning, Image-based Structural Understanding, and Open-world Understanding, encompassing numerical reasoning, structural damage assessment, risk inference, causal analysis, and repair planning. The benchmark is constructed through a systematic pipeline of domain-specific data collection, engineering-oriented captioning, expert annotation, question design, and quality verification. Evaluation of representative state-of-the-art VLMs reveals substantial limitations in domain-specific reasoning, with Sequential Reasoning emerging as a major bottleneck, particularly when models must infer numerical relationships from sparse structured observations. Further experiments show that numerical priors and external knowledge can alleviate these limitations, although their effectiveness depends on the model's ability to reliably interpret and use the available evidence. CivilBench provides a capability-oriented testbed for identifying reasoning limitations and evaluating progress toward reliable AI-assisted engineering decision-making. The benchmark will be publicly available at https://huggingface.co/datasets/traceuryuan/CivilBench.
#04Nov 16, 2026
cs.CV
MetricHand: Geometry-Aware Flow Matching for 4D Hand Reconstruction in Metric World Space
Yibo Bai, Likui Zhang, Jinglei Zhang and 4 more
Egocentric hand motion provides physical motion supervision for dexterous manipulation and vision-language-action pretraining. Reliable use of this supervision requires accurate and temporally coherent 4D hand reconstruction in metric world space. Existing methods typically lift framewise estimates produced under weak perspective into world coordinates. By ignoring joint-dependent depth variation at close range, this approximation couples errors in translation, orientation, and articulation that persist after world lifting and worsen under occlusion. We present MetricHand, a geometry-aware flow matching framework for 4D hand reconstruction in metric world space. MetricHand encodes depth priors from hand-object interaction regions into joint-associated representations that preserve local spatial structure and metric scale. Fused with visual features, this geometry guides coarse metric estimation and conditions every step of temporal generation. The model jointly recovers world-space translation, global orientation, MANO articulation, and shape, while structural priors and temporal context support estimation under partial occlusion. MetricHand reduces camera-space MPJPE by 49.3% and 52.9%, absolute world-space MPJPE by 28.7% and 21.7%, and wrist localization error by 48.5% and 37.9% relative to HaWoR, respectively. World-space acceleration error decreases by 44.2% and 64.3%, demonstrating improved temporal fidelity alongside more accurate absolute localization.
#05Nov 16, 2026
cs.CV
A Study on Performance Enhancement of a Webcam-Based Real-Time Eye-Tracking System
Minuk Jung, Kyeonghun Kim, Sooyeon Yoon
This study proposes a real-time gaze-tracking system that uses a standard RGB webcam to estimate screen coordinates. The system combines ETH-XGaze-based gaze estimation with nine-point calibration and polynomial ridge regression. To reduce errors caused by changes in head position, it also applies a conditional residual correction derived from head and pose features extracted with RTMPose. Whether this correction is used depends on the reliability of the detected features and the characteristics of the calibration data. In a preliminary experiment, one participant completed trials at 12 screen targets. The resulting mean screen-coordinate error was 204.24 pixels, which was 35.83% lower than that of xgaze-js and 57.69% lower than that of gaze_tracker. The system produced valid gaze estimates for all 12 targets. These preliminary findings demonstrate the feasibility of combining user-specific calibration with conditional pose-based correction for webcam gaze tracking. Because the evaluation involved only one participant, further experiments with more participants and a wider range of postures are required.