I2V-GS: Infrastructure-to-Vehicle View Transformation with Gaussian Splatting for Autonomous Driving Data Generation

• URL: http://arxiv.org/abs/2507.23683v1
• Date: Thu, 31 Jul 2025 15:59:16 GMT
• Title: I2V-GS: Infrastructure-to-Vehicle View Transformation with Gaussian Splatting for Autonomous Driving Data Generation
• Authors: Jialei Chen, Wuhao Xu, Sipeng He, Baoru Huang, Dongchun Ren,
• Abstract summary: We introduce a novel method, I2V-GS, to transfer the Infrastructure view To the Vehicle view with Gaussian Splatting.<n>We also introduce RoadSight, a multi-modality, multi-view dataset from real scenarios in infrastructure views.<n>I2V-GS significantly improves quality under vehicle view, outperforming StreetGaussian in NTA-Iou, NTL-Iou, and FID by 45.7%, 34.2%, and 14.9%, respectively.
• Score: 4.041586891110227
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Vast and high-quality data are essential for end-to-end autonomous driving systems. However, current driving data is mainly collected by vehicles, which is expensive and inefficient. A potential solution lies in synthesizing data from real-world images. Recent advancements in 3D reconstruction demonstrate photorealistic novel view synthesis, highlighting the potential of generating driving data from images captured on the road. This paper introduces a novel method, I2V-GS, to transfer the Infrastructure view To the Vehicle view with Gaussian Splatting. Reconstruction from sparse infrastructure viewpoints and rendering under large view transformations is a challenging problem. We adopt the adaptive depth warp to generate dense training views. To further expand the range of views, we employ a cascade strategy to inpaint warped images, which also ensures inpainting content is consistent across views. To further ensure the reliability of the diffusion model, we utilize the cross-view information to perform a confidenceguided optimization. Moreover, we introduce RoadSight, a multi-modality, multi-view dataset from real scenarios in infrastructure views. To our knowledge, I2V-GS is the first framework to generate autonomous driving datasets with infrastructure-vehicle view transformation. Experimental results demonstrate that I2V-GS significantly improves synthesis quality under vehicle view, outperforming StreetGaussian in NTA-Iou, NTL-Iou, and FID by 45.7%, 34.2%, and 14.9%, respectively.

Related papers

• BRUM: Robust 3D Vehicle Reconstruction from 360 Sparse Images [21.811586185200706]
  This paper addresses the challenge of reconstructing vehicles from sparse-view inputs.<n>We leverage depth maps and a robust pose estimation architecture to synthesize novel views.<n>We present a novel dataset featuring both synthetic and real-world public transportation vehicles.
  arXiv  Detail & Related papers  (2025-07-16T10:04:35Z)
• MTGS: Multi-Traversal Gaussian Splatting [51.22657444433942]
  Multi-traversal data provides multiple viewpoints for scene reconstruction within a road block.<n>We propose Multi-Traversal Gaussian Splatting (MTGS), a novel approach that reconstructs high-quality driving scenes from arbitrarily collected multi-traversal data.<n>Our results demonstrate that MTGS improves LPIPS by 23.5% and geometry accuracy by 46.3% compared to single-traversal baselines.
  arXiv  Detail & Related papers  (2025-03-16T15:46:12Z)
• LargeAD: Large-Scale Cross-Sensor Data Pretraining for Autonomous Driving [52.83707400688378]
  LargeAD is a versatile and scalable framework designed for large-scale 3D pretraining across diverse real-world driving datasets.<n>Our framework leverages VFMs to extract semantically rich superpixels from 2D images, which are aligned with LiDAR point clouds to generate high-quality contrastive samples.<n>Our approach delivers significant performance improvements over state-of-the-art methods in both linear probing and fine-tuning tasks for both LiDAR-based segmentation and object detection.
  arXiv  Detail & Related papers  (2025-01-07T18:59:59Z)
• V2X-VLM: End-to-End V2X Cooperative Autonomous Driving Through Large Vision-Language Models [13.716889927164383]
  Vehicle-to-everything (V2X) cooperation has emerged as a promising paradigm to overcome the perception limitations of classical autonomous driving.<n>This paper introduces V2X-VLM, a novel end-to-end (E2E) cooperative autonomous driving framework based on vision-language models (VLMs)<n>V2X-VLM integrates multiperspective camera views from vehicles and infrastructure with text-based scene descriptions to enable a more comprehensive understanding of driving environments.
  arXiv  Detail & Related papers  (2024-08-17T16:42:13Z)
• Learning Lane Graphs from Aerial Imagery Using Transformers [7.718401895021425]
  This work introduces a novel approach to generating successor lane graphs from aerial imagery. We frame successor lane graphs as a collection of maximal length paths and predict them using a Detection Transformer (DETR) architecture. We demonstrate the efficacy of our method through extensive experiments on the diverse and large-scale UrbanLaneGraph dataset.
  arXiv  Detail & Related papers  (2024-07-08T07:42:32Z)
• Leveraging Driver Field-of-View for Multimodal Ego-Trajectory Prediction [69.29802752614677]
  RouteFormer is a novel ego-trajectory prediction network combining GPS data, environmental context, and the driver's field-of-view.<n>To tackle data scarcity and enhance diversity, we introduce GEM, a dataset of urban driving scenarios enriched with synchronized driver field-of-view and gaze data.
  arXiv  Detail & Related papers  (2023-12-13T23:06:30Z)
• Deep Perspective Transformation Based Vehicle Localization on Bird's Eye View [0.49747156441456597]
  Traditional approaches rely on installing multiple sensors to simulate the environment. We propose an alternative solution by generating a top-down representation of the scene. We present an architecture that transforms perspective view RGB images into bird's-eye-view maps with segmented surrounding vehicles.
  arXiv  Detail & Related papers  (2023-11-12T10:16:42Z)
• Street-View Image Generation from a Bird's-Eye View Layout [95.36869800896335]
  Bird's-Eye View (BEV) Perception has received increasing attention in recent years. Data-driven simulation for autonomous driving has been a focal point of recent research. We propose BEVGen, a conditional generative model that synthesizes realistic and spatially consistent surrounding images.
  arXiv  Detail & Related papers  (2023-01-11T18:39:34Z)
• Efficient Self-supervised Vision Transformers for Representation Learning [86.57557009109411]
  We show that multi-stage architectures with sparse self-attentions can significantly reduce modeling complexity. We propose a new pre-training task of region matching which allows the model to capture fine-grained region dependencies. Our results show that combining the two techniques, EsViT achieves 81.3% top-1 on the ImageNet linear probe evaluation.
  arXiv  Detail & Related papers  (2021-06-17T19:57:33Z)
• Fine-Grained Vehicle Perception via 3D Part-Guided Visual Data Augmentation [77.60050239225086]
  We propose an effective training data generation process by fitting a 3D car model with dynamic parts to vehicles in real images. Our approach is fully automatic without any human interaction. We present a multi-task network for VUS parsing and a multi-stream network for VHI parsing.
  arXiv  Detail & Related papers  (2020-12-15T03:03:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.