StutterStitcher
Requirements
- Python 3
- OpenCV 3 (+contrib)
- NumPy
- Matplotlib (for display and debugging)
Usage
usage: stutterstitcher.py [-h] [-n NUM_FOREGROUND] [-r RATE] [-s SCALE]
[-o OUTPUT] [--show] [-v]
data
positional arguments:
data Path to image folder or video file
optional arguments:
-h, --help show this help message and exit
-n NUM_FOREGROUND, --num-foreground NUM_FOREGROUND
Number of foreground objects in scene (default:
autodetect)
-r RATE, --rate RATE Frame resampling rate (e.g. 3 means keep 1 in 3
frames) (default: autodetect)
-s SCALE, --scale SCALE
Scale input images by factor (default: 0.5)
-o OUTPUT, --output OUTPUT
Output filename (default: out.png)
--show Display results
-v, --verbose Display intermediate results and verbose output
Rate is really the most important parameter to set, and should generally be 3-4 for image bursts, or 10-25 for video (these are rough estimates)
Todo
- Foreground object detection
- Sampling rate estimation