Research map / Image generation / Super-resolution
BurstM: Deep Burst Multi-scale SR using Fourier Space with Optical Flow
EungGu Kang, Byeonghun Lee, Sunghoon Im, Kyong Hwan Jin
Read the paperCheck 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 |
|---|---|---|---|
| add Fourier mapping to the INR decoder (ablation) | BurstM · SyntheticBurst (x4) | PSNR (dB): 39.82 -> 41.74 | Improved |
| use FNet instead of SpyNet as optical flow estimator (ablation) | BurstM · SyntheticBurst (x4) | PSNR (dB): 40.85 -> 41.74; 14.3 ms -> 11.6 ms | Improved |
| train on x2, x3, x4 together instead of x4 only (ablation) | BurstM · SyntheticBurst x2 | PSNR (dB): 33.36 -> 44.77 (x2); 42.26 -> 41.74 (x4) | Improved |
| train with continuous floating scale factors instead of integer scales (ablation) | BurstM · SyntheticBurst (x4) | PSNR (dB): 41.74 -> 39.44 | Hurt |
| predict multiple scales directly instead of x4 SR followed by bicubic downsampling | BurstM vs baselines · SyntheticBurst (x2, x3) | PSNR (dB): 41.25 -> 46.01 (x2) | Improved |
| optical flow alignment instead of multi-layer DCN | BurstM · BurstSR (x4) | PSNR (dB): 48.95 -> 49.12 | Improved |
| BurstM at x4 on SyntheticBurst versus transformer-based BSRT-Large | BurstM · SyntheticBurst (x4) | PSNR (dB): 43.62 vs 42.87 | Hurt |
| burst SR with optical flow and Fourier features is faster at inference | BurstM · BurstSR (x4) | inference time (ms): 20.8 -> 11.6 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Burstormer — Burstormer: Burst Image Restoration and Enhancement Transformer (CVPR 2023) · better on 2 of 2 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
- QMambaBSR: Burst Image Super-Resolution with Query State Space Model (CVPR 2025 · similarity 0.87)
- WaveMixSR-V2: Enhancing Super-resolution with Higher Efficiency (Student Abstract) (AAAI 2025 · similarity 0.86)
- BF-STVSR: B-Splines and Fourier---Best Friends for High Fidelity Spatial-Temporal Video Super-Resolution (CVPR 2025 · similarity 0.86)
- Sequence Matters: Harnessing Video Models in 3D Super-Resolution (AAAI 2025 · similarity 0.85)
- Continuous Optical Zooming: A Benchmark for Arbitrary-Scale Image Super-Resolution in Real World (CVPR 2024 · similarity 0.85)
- SAVSR: Arbitrary-Scale Video Super-Resolution via a Learned Scale-Adaptive Network (AAAI 2024 · similarity 0.85)
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.
