引導式測驗 · 中等 等級
如何使用 AI 稳定抖动影片(測驗)
平衡穩定性與裁剪、扭曲、重建邊界、運動偽影、延遲以及保留原始素材的需要。
問題 1 的 8
Why can stronger stabilization reduce the visible field of view?
預覽每個問題8 問題
當您參加測驗時,就會出現答案和解釋。
- Why can stronger stabilization reduce the visible field of view?
- The transform may need cropping or zooming to fill shifted edges
- The camera’s sensor becomes smaller
- Stabilization adds a new lens
- The original frame rate is reduced automatically
- A stabilized building’s straight lines bend during a fast pan. What may be happening?
- The building moved with the camera
- Motion estimation is struggling with rolling shutter or rapid movement
- The output resolution increased
- The clip gained an extra exposure
- A tool fills missing borders with synthesized detail. What should an editor remember?
- They prove the camera saw beyond the frame
- They restore the exact original scene
- Those pixels are inferred, not captured evidence
- They cannot contain artifacts
- A documentary editor stabilizes an interview. What should be preserved?
- The model’s name without source media
- Only a compressed final export
- A screenshot of the strongest correction
- The original clip and a reversible stabilized version
- A clip has motion blur and missed focus. What can stabilization fix?
- It can guarantee sharp focus
- It cannot restore those missing details
- It can recover every blurred frame
- It can replace the camera angle
- How should an editor decide whether stabilization looks natural?
- Measure only file size
- Inspect only one still frame
- Check motion and artifacts at normal playback speed
- Use the highest strength setting
- Why test real-time stabilization on the actual setup?
- Latency and processing depend on camera and capture conditions
- Post-production never uses models
- Lighting affects only audio
- All cameras have identical frame rates
- A stabilized shot is intended as evidence in a factual video. What should the editor avoid?
- Using a mild correction
- Preserving the original file
- Comparing both versions
- Treating synthesized borders as captured scene evidence