Get hands-on experience with 20+ free Google Cloud products and $300 in free credit for new customers.

timeSegments vs timeSegmentAnnotations

timeSegments vs timeSegmentAnnotations

Can anyone explain what's the difference between these 2 fields described here? https://storage.cloud.google.com/google-cloud-aiplatform/schema/dataset/ioformat/video_action_recogn...

why would I want to tag timeSegments? what's the objective of this? associate a label to a time segment?

 

 

0 1 243
1 REPLY 1

timeSegments is a parameter that should be included to your schema's YAML and JSONL files so that it can be used to determine the timestamps of the operations, not something that should be made from the terminal. I.e., a section of the film that contains a number of actions or annotations. And when you import data into your dataset later, you add those files.

However, I checked timeSegmentAnnotations and found that although I set the endTime to be different from startTime, the action is only labeled on the startTime, so it doesn't seem to be that useful (hence, one frame).