TrajFusionNet+ is a transformer-based model that predicts pedestrian crossing intention by fusing sequential trajectory data, visual trajectory overlays, and graph-based scene context. It extends the earlier TrajFusionNet with three attention modules—Sequence, Visual, and Graph—to capture temporal, visual, and relational cues. The model outperforms state‑of‑the‑art methods on the PIE and JAAD datasets and shows better generalization under a joint‑training, separate‑evaluation protocol.
By Fran\c{c}ois G. Landry, Moulay A. Akhloufi
MTF‑Net is a Multi‑Modal Temporal Feature Fusion Network that jointly models kinematic, appearance, and contextual cues for pedestrian intention prediction. It fuses four modalities—bounding‑box dynamics, human pose keypoints, local context, and scene‑level semantics—within a recurrent framework enhanced by gated linear units (GLUs) and an attention‑guided fusion head. Evaluations on the PIE and JAAD benchmarks show that MTF‑Net outperforms recent transformer‑ and graph‑based models, achieving up to 0.95 AUC on PIE and 0.94 AUC on JAAD while maintaining real‑time performance.
By Md Mahfuzur Rahman, Pengzhan Zhou, A. F. M. Abdun Noor, Md Imam Ahasan, Md Mustafizur Rahman, Fang Qu
arXiv:2604. 07126v2 Announce Type: replace-cross Abstract: Predicting vehicle trajectories plays an important role in autonomous driving, transportation safety analysis, traffic operations, etc.
By Diyi Liu, Zihan Niu, Tu Xu, Xingchen Zhang, Lishan Sun
arXiv:2606. 18824v1 Announce Type: cross Abstract: Pedestrian trajectory prediction from an ego-centric camera is challenging since it depends on complex interactions with vehicles and scene context, as well as the intention of the pedestrian.
By Yuxuan Xie, Nicolas Pugeault, Chongfeng Wei, Hubert P. H. Shum, Edmond S. L. Ho
The paper introduces the Multi-Context Fusion Transformer (MFT), a model that predicts pedestrian crossing intentions in urban settings by integrating four types of contextual information—pedestrian behavior, environment, localization, and vehicle motion—through a progressive fusion strategy. MFT uses intra-context attention for reciprocal interactions within each context, cross-context attention to combine these contexts into a global representation, and guided attention mechanisms to refine both context tokens and the global token. Experiments on JAADbeh, JAADall, and PIE datasets show MFT outperforms existing methods with accuracies of 73%, 93%, and 90% respectively, and ablation studies confirm the importance of each network component and input context.
By Yuanzhe Li, Hang Zhong, Steffen M\"uller
arXiv:2609.23507v1 Announce Type: new
Abstract: The increasing reliance on mobile phones has made phone-induced pedestrian distraction increasingly prevalent. Activities such as texting, watching vid...
By Yuanzhe Li, Hounian Liu, Xiaotong Chang, Yidi Huang
arXiv:2608. 12187v1 Announce Type: cross Abstract: Transformer-based methods have achieved strong performance in monocular 3D human pose estimation, but most existing approaches organise spatial and temporal reasoning as separate stages, which may weaken unified spatial-temporal interdependencies inherent in human motion and compress frame-level structural information before temporal modelling.
By Ruochen Li, Shuang Chen, Wenke E, Farshad Arvin, Amir Atapour-Abarghouei
arXiv:2608. 15929v1 Announce Type: new Abstract: Pedestrian path prediction is crucial for enhancing the safety of autonomous vehicles and advanced driver-assistance systems.
By \v{S}imon Sukup, Ariyan Bighashdel, Pavol Jancura
arXiv:2607. 20505v1 Announce Type: cross Abstract: Surrogate safety measures (SSMs) enable proactive traffic safety assessment, but many existing methods evaluate pairwise interactions independently or flatten multi-agent scenes into fixed feature vectors, limiting their ability to represent heterogeneous interaction structure and evolving scene-level risk.
By Md Monzurul Islam, Subasish Das
The paper presents an end‑to‑end system that converts driving footage into dynamic vision sensor (DVS) event streams, augments training with simulated DVS data, and trains a convolutional spiking neural network (Conv‑SNN) to classify pedestrian crossing intent as crossing or non‑crossing. The Conv‑SNN, trained with a class‑balanced loss and surrogate‑gradient learning, achieves high accuracy and F1 scores on JAAD and CARLA DVS datasets, outperforming or matching prior frame‑based methods while operating on sparse temporal representations. The study details architectural choices, neuron dynamics, and training protocols, and provides a convergence analysis and domain‑transfer evaluation.
By Henok Teklu, Mustafa Sakhai, Maciej Wielgosz, Matej Mertik
arXiv:2607. 09740v1 Announce Type: new Abstract: Safe motion planning in advanced driver-assistance systems and autonomous vehicles requires an accurate understanding of how the surrounding traffic scene is likely to evolve.
By Joshua Kofi Asamoah, Blessing Agyei Kyem, Eugene Denteh, Armstrong Aboah
SOMA is a new benchmark that tests fine‑grained egocentric body‑motion understanding, featuring 1,632 human‑verified questions across body configuration, motion dynamics, and temporal reasoning. The study evaluates 37 multimodal large language models (MLLMs) and finds that humans achieve 68.5% accuracy while the best MLLM (Qwen3.6‑27B) reaches only 50.0% zero‑shot accuracy. The authors also provide a training dataset and a multimodal baseline to help close the performance gap and highlight common failure modes such as misreading body parts and temporal errors.
By Matea Tashkovska, Na\"el Ouerghemi, Haozhe Qi, Alexander Mathis