单次前向传播实现未标定多曝光图像的HDR三维重建。
InstantHDR: Single-forward Gaussian Splatting for High Dynamic Range 3D Reconstruction
- 通过几何引导的多曝光融合建模，实现无优化的快速3D HDR重建。
- 在168个渲染场景上训练，相比最优优化方法快700倍以上。
- 适合需要实时重建的移动设备或自动驾驶场景使用。
高动态范围（HDR）新视角合成旨在从多曝光低动态范围（LDR）图像中重建HDR场景。现有方法严重依赖已知相机位姿、良好初始化的稠密点云和耗时的逐场景优化。当前前向传播替代方案因假设曝光不变外观而忽略HDR问题。为填补这一空白，我们提出InstantHDR，一个前向网络，可从未标定的多曝光LDR图像集合中，在单次前向传播中重建3D HDR场景。具体而言，设计了用于多曝光融合的几何引导外观建模，以及通用的场景自适应色调映射元网络。由于缺乏真实HDR场景数据，我们构建了一个预训练数据集HDR-Pretrain，包含168个Blender渲染场景、多种光照类型及多个相机响应函数。全面实验表明，InstantHDR在合成性能上与最先进的基于优化的HDR方法相当，同时在单次前向与后优化设置下分别实现约700倍和20倍的重建速度提升。
原文摘要 · Abstract (English)
High dynamic range (HDR) novel view synthesis (NVS) aims to reconstruct HDR scenes from multi-exposure low dynamic range (LDR) images. Existing HDR pipelines heavily rely on known camera poses, well-initialized dense point clouds, and time-consuming per-scene optimization. Current feed-forward alternatives overlook the HDR problem by assuming exposure-invariant appearance. To bridge this gap, we propose InstantHDR, a feed-forward network that reconstructs 3D HDR scenes from uncalibrated multi-exposure LDR collections in a single forward pass. Specifically, we design a geometry-guided appearance modeling for multi-exposure fusion, and a meta-network for generalizable scene-specific tone mapping. Due to the lack of HDR scene data, we build a pre-training dataset, called HDR-Pretrain, for generalizable feed-forward HDR models, featuring 168 Blender-rendered scenes, diverse lighting types, and multiple camera response functions. Comprehensive experiments show that our InstantHDR delivers comparable synthesis performance to the state-of-the-art optimization-based HDR methods while enjoying $\sim700\times$ and $\sim20\times$ reconstruction speed improvement with our single-forward and post-optimization settings. All code, models, and datasets will be released after the review process.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。