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 ∼700× and ∼20× reconstruction speed improvement with our single-forward and post-optimization settings. All code, models, and datasets will be released after the review process.
@article{arxiv.2603.11298,
title = {InstantHDR: Single-forward Gaussian Splatting for High Dynamic Range 3D Reconstruction},
author = {Dingqiang Ye and Jiacong Xu and Jianglu Ping and Yuxiang Guo and Chao Fan and Vishal M. Patel},
journal= {arXiv preprint arXiv:2603.11298},
year = {2026}
}