# F2HDR: Two-Stage HDR Video Reconstruction via Flow Adapter and Physical Motion Modeling

arXiv: https://arxiv.org/abs/2603.14920 · HTML: https://mlsanity.com/paper/2603.14920/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 42 references that are in the map.
- First posted on arXiv 16 Mar 2026.

## Builds on (most cited first, 22 of 22)

- [Video Enhancement with Task-Oriented Flow](https://mlsanity.com/paper/1711.09078/) (2017)
- [High Speed and High Dynamic Range Video with an Event Camera](https://mlsanity.com/paper/1906.07165/) (2019)
- [SEA-RAFT: Simple, Efficient, Accurate RAFT for Optical Flow](https://mlsanity.com/paper/2405.14793/) (2024)
- [Attention-guided Network for Ghost-free High Dynamic Range Imaging](https://mlsanity.com/paper/1904.10293/) (2019)
- [Deep High Dynamic Range Imaging with Large Foreground Motions](https://mlsanity.com/paper/1711.08937/) (2017)
- [Ghost-free High Dynamic Range Imaging with Context-aware Transformer](https://mlsanity.com/paper/2208.05114/) (2022)
- [MemFlow: Optical Flow Estimation and Prediction with Memory](https://mlsanity.com/paper/2404.04808/) (2024)
- [ADNet: Attention-guided Deformable Convolutional Network for High Dynamic Range Imaging](https://mlsanity.com/paper/2105.10697/) (2021)
- [HDR Video Reconstruction: A Coarse-to-fine Network and A Real-world Benchmark Dataset](https://mlsanity.com/paper/2103.14943/) (2021)
- [Alignment-free HDR Deghosting with Semantics Consistent Transformer](https://mlsanity.com/paper/2305.18135/) (2023)
- [HDRFlow: Real-Time HDR Video Reconstruction with Large Motions](https://mlsanity.com/paper/2403.03447/) (2024)
- [SAFNet: Selective Alignment Fusion Network for Efficient HDR Imaging](https://mlsanity.com/paper/2407.16308/) (2024)
- [UltraFusion: Ultra High Dynamic Imaging using Exposure Fusion](https://mlsanity.com/paper/2501.11515/) (2025)
- [A Unified Framework for Multi-Sensor HDR Video Reconstruction](https://mlsanity.com/paper/1308.4908/) (2013)
- [LAN-HDR: Luminance-based Alignment Network for High Dynamic Range Video Reconstruction](https://mlsanity.com/paper/2308.11116/) (2023)
- [FlexHDR: Modelling Alignment and Exposure Uncertainties for Flexible HDR Imaging](https://mlsanity.com/paper/2201.02625/) (2022)
- [Progressive and Selective Fusion Network for High Dynamic Range Imaging](https://mlsanity.com/paper/2108.08585/) (2021)
- [Towards Real-World HDR Video Reconstruction: A Large-Scale Benchmark Dataset and A Two-Stage Alignment Network](https://mlsanity.com/paper/2405.00244/) (2024)
- [AFUNet: Cross-Iterative Alignment-Fusion Synergy for HDR Reconstruction via Deep Unfolding Paradigm](https://mlsanity.com/paper/2506.23537/) (2025)
- [Exposure Completing for Temporally Consistent Neural High Dynamic Range Video Rendering](https://mlsanity.com/paper/2407.13309/) (2024)
- [HDR Video Reconstruction with a Large Dynamic Dataset in Raw and sRGB Domains](https://mlsanity.com/paper/2304.04773/) (2023)
- [Diffusion-Promoted HDR Video Reconstruction](https://mlsanity.com/paper/2406.08204/) (2024)

## Built on by (most cited first, 2 of 2)

- [Revisiting the Current Frame: Physical-Trace-Guided Network Output Correction for Video Restoration](https://mlsanity.com/paper/2608.09342/) (2026)
- [Double-stream registration with pyramid fusion for HDR video with alternating exposures](https://mlsanity.com/paper/2609.31108/) (2026)

## Abstract (arXiv)

Reconstructing High Dynamic Range (HDR) videos from sequences of alternating-exposure Low Dynamic Range (LDR) frames remains highly challenging, especially under dynamic scenes where cross-exposure inconsistencies and complex motion make inter-frame alignment difficult, leading to ghosting and detail loss. Existing methods often suffer from inaccurate alignment, suboptimal feature aggregation, and degraded reconstruction quality in motion-dominated regions. To address these challenges, we propose F2HDR, a two-stage HDR video reconstruction framework that robustly perceives inter-frame motion and restores fine details in complex dynamic scenarios. The proposed framework integrates a flow adapter that adapts generic optical flow for robust cross-exposure alignment, a physical motion modeling to identify salient motion regions, and a motion-aware refinement network that aggregates complementary information while removing ghosting and noise. Extensive experiments demonstrate that F2HDR achieves state-of-the-art performance on real-world HDR video benchmarks, producing ghost-free and high-fidelity results under large motion and exposure variations.