https://github.com/pvlbzn/avgframe
Reverse of a time lapse videos. Map video to one image.
https://github.com/pvlbzn/avgframe
opencv video-processing
Last synced: 3 months ago
JSON representation
Reverse of a time lapse videos. Map video to one image.
- Host: GitHub
- URL: https://github.com/pvlbzn/avgframe
- Owner: pvlbzn
- Created: 2017-06-16T12:11:23.000Z (about 9 years ago)
- Default Branch: master
- Last Pushed: 2017-06-18T11:03:14.000Z (about 9 years ago)
- Last Synced: 2025-10-28T02:41:40.733Z (8 months ago)
- Topics: opencv, video-processing
- Language: Python
- Homepage:
- Size: 44.2 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Average Frame
Produce one frame from a sequence of frames such that resulting
frame represents an average of a sequence.
## Algorithm
&space;\;&space;/&space;\;&space;|\,&space;\chi\,&space;|)

Where _X_ is a set of frames, x is a particular frame.
## Usage
From terminal:
```
python3 path/to/video/file.mp4 frames_to_skip frames_to_grab
```
From REPL:
```
from src import averanger as a
# Path to your video file
path = 'data/video.mp4'
# Frames to skip
skip = 2000
# Frames to grab
grab = 25
a.averengify(path, skip, grab)
```
## Outcomes
There is no one unique recipe besides basic video analysis. General rules are the following:
* The more video is centered on some object and more this object 'still' the bigger grab value can be taken
* The more video intense (change of the background, absence of central or 'still' object) the smaller grab value should be
#### 170 Frames, Still

#### 5 frames, Action

#### 10 frames, Action

#### 25 frames, Action

#### 50 frames, Action
