SeHDR: Single-Exposure HDR Novel View Synthesis via 3D Gaussian Bracketing

2025-09-23

AI Summary

This work addresses the challenging problem of synthesizing HDR novel views from a single-exposure set of multi-view LDR images—bypassing distortions (e.g., motion blur and misalignment) inherent in conventional multi-exposure approaches. We propose *Bracketed 3D Gaussians*, the first method to jointly estimate scene geometry and linear color distributions across multiple exposure values (EVs) within the 3D Gaussian splatting framework, using only single-exposure inputs; colors are represented via spherical harmonics. To fuse these EV-specific 3D Gaussians robustly in HDR space, we introduce a differentiable Neural Exposure Fusion (NeEF) module. Experiments on both synthetic and real-world datasets demonstrate that our approach significantly outperforms existing HDR novel view synthesis methods, achieving state-of-the-art performance in visual quality, dynamic range recovery, and geometric consistency.

Technology Category

• Computer Vision: 3D Computer Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

This paper presents SeHDR, a novel high dynamic range 3D Gaussian Splatting (HDR-3DGS) approach for generating HDR novel views given multi-view LDR images. Unlike existing methods that typically require the multi-view LDR input images to be captured from different exposures, which are tedious to capture and more likely to suffer from errors (e.g., object motion blurs and calibration/alignment inaccuracies), our approach learns the HDR scene representation from multi-view LDR images of a single exposure. Our key insight to this ill-posed problem is that by first estimating Bracketed 3D Gaussians (i.e., with different exposures) from single-exposure multi-view LDR images, we may then be able to merge these bracketed 3D Gaussians into an HDR scene representation. Specifically, SeHDR first learns base 3D Gaussians from single-exposure LDR inputs, where the spherical harmonics parameterize colors in a linear color space. We then estimate multiple 3D Gaussians with identical geometry but varying linear colors conditioned on exposure manipulations. Finally, we propose the Differentiable Neural Exposure Fusion (NeEF) to integrate the base and estimated 3D Gaussians into HDR Gaussians for novel view rendering. Extensive experiments demonstrate that SeHDR outperforms existing methods as well as carefully designed baselines.

Problem

• Generating HDR novel views from single-exposure LDR images
• Overcoming limitations of multi-exposure capture requirements
• Learning HDR scene representation from limited exposure data

Innovation

• HDR novel views from single-exposure LDR images
• Estimates bracketed 3D Gaussians with varying exposures
• Uses Differentiable Neural Exposure Fusion for HDR integration