Research map / Video / Video generation
Harnessing Meta-Learning for Controllable Full-Frame Video Stabilization
Muhammad Kashif Ali, Eun Woo Im, D Kim, Tae Hyun Kim, Vivek Gupta, Haonan Luo, Tianrui Li
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 targeted (jerk-localized) adaptation instead of random-sequence adaptation (ablation) | DIFRINT · NUS | stability score: 0.8528 -> 0.8620 | Improved |
| use targeted rather than vanilla adaptation (ablation) | DIFRINT · NUS | distortion: 0.9596 -> 0.9354 | Hurt |
| raise number of adapted peak sequences p from 10 to 100 in targeted adaptation (ablation) | DIFRINT + TargetedAdapt · NUS | stability score: 0.8620 -> 0.8680 | Improved |
| raise p from 10 to 100 in targeted adaptation (ablation) | DIFRINT + TargetedAdapt · NUS | distortion: 0.9354 -> 0.9542 | Improved |
| add test-time meta-adaptation (1 pass, 100 sequences) to DIFRINT | DIFRINT · NUS | stability score: 0.7904 -> 0.8428 | Improved |
| add one adaptation pass over all sequences to DIFRINT | DIFRINT · NUS | stability score: 0.7904 -> 0.8786 | Improved |
| add vanilla adaptation over all sequences to DMBVS | DMBVS · NUS | stability score: 0.7372 -> 0.7760 | Improved |
| add targeted adaptation and recurrent formulation to DMBVS | DMBVS · DOFVS | stability score: 0.7577 -> 0.8253 | Improved |
| add targeted adaptation to DIFRINT | DIFRINT · DOFVS | stability score: 0.7481 -> 0.8331 | Improved |
| add targeted adaptation to DIFRINT | DIFRINT · BiT | stability score: 0.7587 -> 0.7962 | Improved |
| add targeted adaptation to DMBVS on DOFVS | DMBVS · DOFVS | distortion: 0.9531 -> 0.9061 | Hurt |
| add targeted adaptation to DIFRINT | DIFRINT · NUS | average persistence: 27.970 -> 36.254 | Improved |
Closest papers by meaning
- Harnessing Meta-Learning for Improving Full-Frame Video Stabilization (CVPR 2024 · similarity 0.97)
- Instant Video Models: Universal Adapters for Stabilizing Image-Based Networks (NeurIPS 2025 · similarity 0.87)
- Training-free Camera Control for Video Generation (ICLR 2025 · similarity 0.86)
- Ctrl-Adapter: An Efficient and Versatile Framework for Adapting Diverse Controls to Any Diffusion Model (ICLR 2025 · similarity 0.86)
- Rebalancing Reference Frame Dominance to Improve Motion in Image-to-Video Models (NeurIPS 2026 · similarity 0.85)
- AC3D: Analyzing and Improving 3D Camera Control in Video Diffusion Transformers (CVPR 2025 · 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.
