[
https://issues.apache.org/jira/browse/TIKA-2322?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15992436#comment-15992436
]
ASF GitHub Bot commented on TIKA-2322:
--------------------------------------
ThejanW commented on issue #168: fix for TIKA-2322 contributed by
msha...@usc.edu
URL: https://github.com/apache/tika/pull/168#issuecomment-298511180
@smadha Could you also add this fact to wiki, when running the container, it
always downloads model files. If a user needs to keep those models, without
downloading them again and again, the could commit those changes to the
container. Once committed those changes, running the container again won't
download those models. This is how to do it,
1. First run the container by this command,
**docker run -p 8764:8764 -it inception-video-rest-tika**
2. Open another terminal, without closing the terminal which is server is
running, then type this command,
**docker ps -l**
3. User will see the immediate container of inception-video-rest-tika and
it's container id, make note of the container id.
4. Then by this command, changes will be committed to
inception-video-rest-tika container,
**docker commit inception-video-rest-tika container_id**
5. To check if it's worked, stop and start the inception-video-rest-tika
container again, it won't download models.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
> Video labeling using existing ObjectRecognition
> -----------------------------------------------
>
> Key: TIKA-2322
> URL: https://issues.apache.org/jira/browse/TIKA-2322
> Project: Tika
> Issue Type: Improvement
> Components: parser
> Reporter: Madhav Sharan
> Assignee: Chris A. Mattmann
> Labels: memex
> Fix For: 1.15
>
>
> Currently TIKA supports ObjectRecognition in Images. I am proposing to extend
> this to support videos.
> Idea is -
> 1. Extract frames from video and run IncV3 to get labels for these frames.
> 2. We average confidence scores of same labels for each frame.
> 3. Return results in sorted order of confidence score.
> I am writing code for different modes of frame extractions -
> 1. Extract center image.
> 2. Extract frames after every fixed interval.
> 3. Extract N frames equally divided across video.
> We used this approach in [0]. Code in [1]
> [0] https://github.com/USCDataScience/hadoop-pot
> [1] https://github.com/USCDataScience/video-recognition
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
