The Video Event Segmentation Tool (VEST) is an open-source software for annotating temporal events in video clips.
You can annotate event "segments" with start & end times or event "points" that indicate only the start time of an event.
More detailed tutorial is available here.
Easiest way to use VEST is through the online app at https://jinhyuncheong.com/vest/vest.html. Alternatively, you can clone this repo and launch index.html in your browser.
- Load your video file using the [Choose file] button
- Define the name of the event segments and points using the [Add segment buttons] and [Add point buttons]. This will create buttons with your event names on them.
- Click your label buttons to annotate segments or events as you watch your video.
- When you are finished, save your annotations by using the [Download results] button.
VEST saves the segmentations into a json file that can be loaded in most coding languages (e.g. Python, R). Here is a sample code to load the outputs in Python [Gist].
import pandas as pd
def parse_vest_json(f):
"""Parse json output from VEST
Args:
f: path to VEST json output
Returns:
segmentations dataframe
points dataframe
"""
seg_df, pts_df = pd.DataFrame(), pd.DataFrame()
df = pd.read_json(f, lines=True)
seg_cols = [col for col in df.columns if 'segment' in col]
_seg_df = df[seg_cols].dropna().T
for rowix, row in _seg_df.iterrows():
seg_df = pd.concat([seg_df, pd.DataFrame(row.values[0],index=[row.name])])
pts_cols = [col for col in df.columns if 'point' in col]
_pts_df = df[pts_cols].dropna().T
for rowix, row in _pts_df.iterrows():
pts_df = pd.concat([pts_df, pd.DataFrame(row.values[0], index=[row.name])])
return seg_df, pts_df
path_to_file = "/content/segmentation_results.json"
seg_df, pts_df = parse_vest_json(f=path_to_file)
display(seg_df, pts_df)