publications

2026

1. ECCV

   VIPS: Vehicle-Infrastructure Cooperative Planning Benchmark via Pseudo-Simulation

   Hoonhee Cho, Jae-Young Kang, Giwon Lee, Hyemin Yang, Heejun Park, and Kuk-Jin Yoon

   In European Conference on Computer Vision. *Equal contribution, †Corresponding author , 2026

   Abs Bib HTML Code Website

   End-to-end autonomous driving (E2E-AD) in urban environments requires robust decision-making under partial observability and complex multi-agent interactions. Severe occlusions and dense traffic at intersections limit the perception capability of single-agent systems, motivating recent efforts on Vehicle-to-Infrastructure (V2I) cooperation for perception and planning. However, existing evaluation protocols face a fundamental trade-off: open-loop evaluation fails to capture error accumulation and recovery from deviations, while closed-loop evaluation is costly, difficult to scale, and often relies on simulated environments that may suffer from domain gaps. To bridge this gap, we propose VIPS, a benchmark for cooperative autonomous driving in V2I settings based on pseudo-simulation. VIPS extends pseudo-simulation to multi-agent scenarios by integrating vehicle and infrastructure observations and introducing interaction-aware perturbations that approximate realistic traffic dynamics. This enables scalable yet realistic evaluation of robustness and error propagation without full simulation. We further present CoS-V2X, a cooperative planning framework based on sparse representations. CoS-V2X models vehicle-infrastructure interactions using compact features for efficient communication and robust decision-making under heterogeneous observations.

   @inproceedings{cho2026vips,
     title = {VIPS: Vehicle-Infrastructure Cooperative Planning Benchmark via Pseudo-Simulation},
     author = {Cho, Hoonhee and Kang, Jae-Young and Lee, Giwon and Yang, Hyemin and Park, Heejun and Yoon, Kuk-Jin},
     booktitle = {European Conference on Computer Vision},
     year = {2026},
   }

2. arXiv

   HEAT: Heterogeneous End-to-End Autonomous Driving via Trajectory-Guided World Models

   Hoonhee Cho, Giwon Lee, Jae-Young Kang, Hyemin Yang, Heejun Park, and Kuk-Jin Yoon

   arXiv preprint, 2026

   Abs Bib HTML

   End-to-end autonomous driving (E2E-AD) has emerged as a compelling alternative to traditional modular pipelines. While recent approaches achieve strong performance on single-domain datasets, their performance degrades significantly when trained jointly across multiple heterogeneous domains. We propose HEAT, a trajectory-driven learning paradigm that organizes training around planning trajectories, enabling the model to capture domain-invariant representations of driving intent.

   @article{yang2025heat,
     title = {HEAT: Heterogeneous End-to-End Autonomous Driving via Trajectory-Guided World Models},
     author = {Cho, Hoonhee and Lee, Giwon and Kang, Jae-Young and Yang, Hyemin and Park, Heejun and Yoon, Kuk-Jin},
     journal = {arXiv preprint},
     year = {2026},
   }

2025

1. NeurIPS

   VR-Drive: Viewpoint-Robust End-to-End Driving with Feed-Forward 3D Gaussian Splatting

   Hoonhee Cho, Jae-Young Kang, Giwon Lee, Hyemin Yang, Heejun Park, Seokwoo Jung, and Kuk-Jin Yoon

   In Conference on Neural Information Processing Systems. *Equal contribution , 2025

   Abs arXiv Bib HTML PDF Code

   End-to-end autonomous driving (E2E-AD) has emerged as a promising paradigm that unifies perception, prediction, and planning into a holistic, data-driven framework. However, achieving robustness to varying camera viewpoints, a common real-world challenge due to diverse vehicle configurations, remains an open problem. In this work, we propose VR-Drive, a novel E2E-AD framework that addresses viewpoint generalization by jointly learning 3D scene reconstruction as an auxiliary task to enable planning-aware view synthesis. Unlike prior scene-specific synthesis approaches, VR-Drive adopts a feed-forward inference strategy that supports online training-time augmentation from sparse views without additional annotations. To further improve viewpoint consistency, we introduce a viewpoint-mixed memory bank that facilitates temporal interaction across multiple viewpoints and a viewpoint-consistent distillation strategy that transfers knowledge from original to synthesized views. Trained in a fully end-to-end manner, VR-Drive effectively mitigates synthesis-induced noise and improves planning under viewpoint shifts. In addition, we release a new benchmark dataset to evaluate E2E-AD performance under novel camera viewpoints, enabling comprehensive analysis. Our results demonstrate that VR-Drive is a scalable and robust solution for the real-world deployment of end-to-end autonomous driving systems.

   @inproceedings{cho2025vrdrive,
     title = {VR-Drive: Viewpoint-Robust End-to-End Driving with Feed-Forward 3D Gaussian Splatting},
     author = {Cho, Hoonhee and Kang, Jae-Young and Lee, Giwon and Yang, Hyemin and Park, Heejun and Jung, Seokwoo and Yoon, Kuk-Jin},
     booktitle = {Conference on Neural Information Processing Systems},
     year = {2025},
   }