Research map / Image generation / Image restoration
Digital Gimbal: End-to-End Deep Image Stabilization With Learnable Exposure Times
Omer Dahary, Matan Jacoby, Alex M. Bronstein
Check your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use learned non-uniform exposure times instead of a fixed-exposure burst (ablation) | Digital Gimbal · Synthetic test set | PSNR: 28.72 -> 31.42 | Improved |
| Pre-align burst frames with a flow network (ablation) | Digital Gimbal · Synthetic test set | PSNR: 30.62 -> 31.42 | Improved |
| Normalize frames by their exposure time (ablation) | Digital Gimbal · Synthetic test set | PSNR: 30.96 -> 31.42 | Improved |
| Add the annealed alignment loss term (ablation) | Digital Gimbal · Synthetic test set | PSNR: 31.18 -> 31.42 | Improved |
| Use learned exposure schedules and burst reconstruction rather than fixed-exposure KPN denoising | Synthetic test set · Synthetic test set | PSNR: 28.46 -> 31.42 | Improved |
| Use GenGS for low photon counts instead of a simple Gaussian approximation | Digital Gimbal sensor forward model · Low to moderate SNR levels | PSNR: around 0.1 dB | Improved |
Later papers that compared with this method
1 later accepted paper reports Digital-Gimbal as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Dynamic Exposure Burst Image Restoration (CVPR 2026 · better on 3 of 3 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Minimum Latency Deep Online Video Stabilization (ICCV 2023 · similarity 0.85)
- Robust Video Stabilization by Optimization in CNN Weight Space (CVPR 2019 · similarity 0.85)
- Minimum Latency Deep Online Video Stabilization and Its Extensions (TPAMI 2024 · similarity 0.84)
- Harnessing Meta-Learning for Improving Full-Frame Video Stabilization (CVPR 2024 · similarity 0.82)
- AC3D: Analyzing and Improving 3D Camera Control in Video Diffusion Transformers (CVPR 2025 · similarity 0.82)
- Learning to Control Camera Exposure via Reinforcement Learning (CVPR 2024 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
