geowatch.cli.animate_visualizations module¶

geowatch.cli.animate_visualizations.animate_visualizations(viz_dpath, channels=None, video_names=None, frames_per_second=0.7, draw_anns=True, draw_imgs=True, workers=0, zoom_to_tracks=False, verbose=0)[source]¶

Helper that roughly does the same thing as this bash script:

Parameters:

viz_dpath (str) – the path where visualizations were dumped with the coco_visualize_videos script.
zoom_to_tracks (bool) – if specified uses “track” based-logic find paths to animate

Example

>>> # xdoctest: +SKIP
>>> # xdoctest: +REQUIRES(--ffmpeg-test')
>>> import ubelt as ub
>>> dpath = ub.Path.appdir('geowatch/tests/ani_video').delete().ensuredir()
>>> import kwcoco
>>> from geowatch.utils import kwcoco_extensions
>>> dset = kwcoco.CocoDataset.demo('vidshapes2-msi', num_frames=5)
>>> img = dset.dataset['images'][0]
>>> coco_img = dset.coco_image(img['id'])
>>> channel_chunks = list(ub.chunks(coco_img.channels.fuse().parsed, chunksize=3))
>>> channels = ','.join(['|'.join(p) for p in channel_chunks])
>>> kwargs = {
>>>     'src': dset.fpath,
>>>     'viz_dpath': dpath,
>>>     'space': 'video',
>>>     'channels': channels,
>>>     'zoom_to_tracks': False,
>>> }
>>> from geowatch.cli.coco_visualize_videos import main
>>> cmdline = False
>>> main(cmdline=cmdline, **kwargs)
>>> viz_dpath = dpath
>>> channels = None
>>> video_names = None
>>> frame_per_second = 0.7
>>> from geowatch.cli.animate_visualizations import *  # NOQA
>>> animate_visualizations(viz_dpath, verbose=1, workers=0)