InstantHDR: Single-forward Gaussian Splatting for High Dynamic Range 3D Reconstruction
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917350875332608 |
|---|---|
| author | Ye, Dingqiang Xu, Jiacong Ping, Jianglu Guo, Yuxiang Fan, Chao Patel, Vishal M. |
| author_facet | Ye, Dingqiang Xu, Jiacong Ping, Jianglu Guo, Yuxiang Fan, Chao Patel, Vishal M. |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_11298 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | InstantHDR: Single-forward Gaussian Splatting for High Dynamic Range 3D Reconstruction Ye, Dingqiang Xu, Jiacong Ping, Jianglu Guo, Yuxiang Fan, Chao Patel, Vishal M. Computer Vision and Pattern Recognition 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. |
| title | InstantHDR: Single-forward Gaussian Splatting for High Dynamic Range 3D Reconstruction |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.11298 |