1. rcrnstn/pyffstream
  2. README.md
Browse code

Add autozoom usage example

Robert Cranston authored on 04/06/2020 17:59:48
Showing 1 changed files
README.md
History View file @ fe601bd
... ... 
@@ -81,6 +81,14 @@ A number of hepler functions are also supplied:
81 81 
 [`PIL.Image.resize`]: https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image.resize
82 82 
 [`cv2.putText`]: https://docs.opencv.org/master/d6/d6e/group__imgproc__draw.html#ga5126f47f883d730f633d74f07456c576
83 83
84 
+### Examples
85 
+
86 
+Projects using `pyffstream`:
87 
+
88 
+-   [`autozoom`][]: Automatically zoom in on video content of interest.
89 
+
90 
+[`autozoom`]: https://git.rcrnstn.net/rcrnstn/autozoom
91 
+
84 92 
 ## Installation
85 93
86 94 
 ### With `pip`
Browse code

Add implementation

Robert Cranston authored on 04/06/2020 17:59:34
Showing 1 changed files
README.md
History View file @ 0be6a56
... ... 
@@ -2,6 +2,107 @@
2 2
3 3 
 A Python library to ease processing of video frames.
4 4
5 
+`pyffstream` is mostly a wrapper around external [FFmpeg][] processes (via
6 
+[`ffmpeg-python`][]) intended to be used with [OpenCV][].
7 
+
8 
+[FFmpeg]: https://ffmpeg.org
9 
+[`ffmpeg-python`]: https://github.com/kkroening/ffmpeg-python
10 
+[OpenCV]: https://opencv.org
11 
+
12 
+## Usage
13 
+
14 
+`pyffstream` provides your application with a command line interface (CLI) and
15 
+handles reading and writing video files (with the original audio intact),
16 
+filtering the frames through a callback that you define. The API consists
17 
+mainly of the single function
18 
+
19 
+```python
20 
+run(description, process, init=None, args_pre=None, args_post=None)
21 
+```
22 
+
23 
+that takes the application description to present in the CLI and some
24 
+callbacks, all but one being optional:
25 
+
26 
+-   `process(args, state, frame, frame_num)`:
27 
+
28 
+    Takes:
29 
+
30 
+    -   `args`: [`argparse.Namespace`][] containing parsed command line
31 
+        arguments.
32 
+    -   `state`: arbitrary object returned by `init` (see below).
33 
+    -   `frame`: [`numpy.array`][] with shape
34 
+        `(args.working_width, args.working_height, 3)` containing RGB data to
35 
+        be processed.
36 
+    -   `frame_num`: integer in the range [`args.frame_start`,
37 
+        `args.frame_end`) representing the number of the current frame.
38 
+
39 
+    Returns:
40 
+
41 
+    -   `output_frame`: [`numpy.array`][] with shape
42 
+        `(args.output_width, args.output_height, 3)` containing RGB output.
43 
+    -   `debug_frame`: [`numpy.array`][] with shape
44 
+        `(args.working_width, args.working_height, 3)` containing RGB debug
45 
+        output.
46 
+
47 
+-   `init(args)`:
48 
+
49 
+    Takes:
50 
+
51 
+    -   `args`: [`argparse.Namespace`][] containing parsed command line
52 
+        arguments.
53 
+
54 
+    Returns:
55 
+
56 
+    -   `state`: arbitrary object passed to `process` (see above).
57 
+
58 
+-   `args_pre(parser)`, `args_post(parser)`:
59 
+
60 
+    Takes:
61 
+
62 
+    -   `args`: [`argparse.ArgumentParser`][] before (for `args_pre`) or after
63 
+        (for `args_post`) being filled with `pyffstream` arguments.
64 
+
65 
+    Returns nothing.
66 
+
67 
+A number of hepler functions are also supplied:
68 
+
69 
+-   `resize(image, width, height, resample)`: wrapper around
70 
+    [`PIL.Image.resize`][].
71 
+-   `text(image, text, x, y, thickness, font=cv2.FONT_HERSHEY_SIMPLEX)`:
72 
+    wrapper around [`cv2.putText`][] that automatically adjusts text placement
73 
+    and scaling.
74 
+-   `fix_rect(args, x, y, w, h)`: forces a rectangle to aspect ratio
75 
+    `args.output_aspect` and to be inside `(0, 0, args.working_width,
76 
+    args.working_height)`.
77 
+
78 
+[`argparse.Namespace`]: https://docs.python.org/3/library/argparse.html#argparse.Namespace
79 
+[`argparse.ArgumentParser`]: https://docs.python.org/3/library/argparse.html#argparse.ArgumentParser
80 
+[`numpy.array`]: https://numpy.org/doc/stable/reference/generated/numpy.array.html
81 
+[`PIL.Image.resize`]: https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image.resize
82 
+[`cv2.putText`]: https://docs.opencv.org/master/d6/d6e/group__imgproc__draw.html#ga5126f47f883d730f633d74f07456c576
83 
+
84 
+## Installation
85 
+
86 
+### With `pip`
87 
+
88 
+```sh
89 
+python3 -m pip install git+https://git.rcrnstn.net/rcrnstn/pyffstream
90 
+```
91 
+
92 
+### In `setuptools` `setup.py`
93 
+
94 
+```python
95 
+from setuptools import setup
96 
+
97 
+setup(
98 
+    # ...
99 
+    install_requires=[
100 
+        'pyffstream @ git+https://git.rcrnstn.net/rcrnstn/pyffstream',
101 
+    ],
102 
+    # ...
103 
+)
104 
+```
105 
+
5 106 
 ## License
6 107
7 108 
 Licensed under the [ISC license][], see the [`LICENSE`](LICENSE) file.
Browse code

Add license

Robert Cranston authored on 04/06/2020 17:57:55
Showing 1 changed files
README.md
History View file @ c54099c
... ... 
@@ -1,3 +1,9 @@
1 1 
 # `pyffstream`
2 2
3 3 
 A Python library to ease processing of video frames.
4 
+
5 
+## License
6 
+
7 
+Licensed under the [ISC license][], see the [`LICENSE`](LICENSE) file.
8 
+
9 
+[ISC license]: https://en.wikipedia.org/wiki/ISC_license
Browse code

Add readme

Robert Cranston authored on 04/06/2020 17:57:36
Showing 1 changed files
README.md
History View file @ e660c61
1 1 
new file mode 100644
... ... 
@@ -0,0 +1,3 @@
1 
+# `pyffstream`
2 
+
3 
+A Python library to ease processing of video frames.