[
https://issues.apache.org/jira/browse/TIKA-2322?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15966870#comment-15966870
]
ASF GitHub Bot commented on TIKA-2322:
--------------------------------------
smadha opened a new pull request #168: fix for TIKA-2322 contributed by
[email protected]
URL: https://github.com/apache/tika/pull/168
Instructions
- Install OpneCV with python support. on mac it can be done by `brew
install opencv` and adding cv2 to python path
- Bring up inception server as per Tika and Vision wiki
- java -jar tika-app/target/tika-app-1.14.jar
--config=tika-parsers/src/test/resources/org/apache/tika/parser/recognit
ion/tika-config-tflow-video-rest.xml
./tika-parsers/src/test/resources/test-documents/testVideoMp4.mp4
- Sample video downloaded from http://www.sample-videos.com/
output from Tika
```
$ java -jar tika-app/target/tika-app-1.14.jar
--config=tika-parsers/src/test/resources/org/apache/tika/parser/recognit
ion/tika-config-tflow-video-rest.xml
./tika-parsesources/test-documents/testVideoMp4.mp4
INFO Available = true, API Status = HTTP/1.0 200 OK
INFO minConfidence = 0.015, topN=4
INFO Recogniser =
org.apache.tika.parser.recognition.tf.TensorflowRESTVideoRecogniser
INFO Recogniser Available = true
<?xml version="1.0" encoding="UTF-8"?><html
xmlns="http://www.w3.org/1999/xhtml";>
<head>
<meta name="org.apache.tika.parser.recognition.object.rec.impl"
content="org.apache.tika.parser.recognition.tf.TensorflowRESTVideoRecogn
iser"/>
<meta name="X-Parsed-By"
content="org.apache.tika.parser.CompositeParser"/>
<meta name="X-Parsed-By"
content="org.apache.tika.parser.recognition.ObjectRecognitionParser"/>
<meta name="resourceName" content="testVideoMp4.mp4"/>
<meta name="Content-Length" content="1053651"/>
<meta name="OBJECT" content="dugong, Dugong dugon (0.15187)"/>
<meta name="OBJECT" content="television, television system (0.08859)"/>
<meta name="OBJECT" content="ice bear, polar bear, Ursus Maritimus,
Thalarctos maritimus (0.06996)"/>
<meta name="OBJECT" content="triceratops (0.05711)"/>
<meta name="Content-Type" content="video/mp4"/>
<title/>
</head>
<body><ol id="objects"> <li id="dugong, Dugong dugon"> dugong, Dugong
dugon [en](confidence = 0.151870 )</li>
<li id="television, television system"> television, television system
[en](confidence = 0.088589 )</li>
<li id="ice bear, polar bear, Ursus Maritimus, Thalarctos maritimus">
ice bear, polar bear, Ursus Maritimus, Thalarctos maritimus
[en](confidence = 0.069961 )</li>
<li id="triceratops"> triceratops [en](confidence = 0.057112 )</li>
</ol>
</body></html>LMC-032212:tika sharan$
```
----------------------------------------------------------------
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:
[email protected]
> 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)
