KO
|
EN
gitlite — search
Search
#javascript
#python
#hacktoberfest
#react
#ai
#typescript
#llm
#go
#golang
#android
#machine-learning
#rust
#deep-learning
#linux
video-recognition
★ 8
Open GitHub ↗
No description available.
Download README (.md)
Explore Similar Repositories
8dot3-brute
:
8dot3-brute is a script to brute force Windows 8.3 filenames and directories in web servers
ex_pre_commit_hook
:
pre commit hook for elixir project which handles compile / credo / test / docs
ILFPT
:
Integrated Learning Framework for Pedestrian Tracking
plasma-kde-connect-skill
:
This skill integrates KDE Connect with Mycroft to enable users to control their Phones using voice commands.
Love-Strawberries-openCV
:
Object Detection Using OpenCV+Python (No Deep Learning)
// repository documentation
Was this content helpful?
★ 0
(0 ratings)
Select Rating:
★
★
★
★
★
Submit Feedback
Recent Feedback
×
Download README
Do you want to download the
README.md
file for
video-recognition
?
Download (.md)
# video-recognition This repo takes advantage of Tika and TensorFlow imagenet to understand a video file. It selects frame of video and saves it as image file. It then uses Inception-V3 model from Tensorflow integrated with Tika to generate description of that image. How to use? 1. `cd src` 2. `python form_label_using_tika.py` ``` Usage - python form_label_using_tika.py [v/i] <path/to/(video/image)/dir> <path/to/output/dir> [v/i] -> v from video, i for images <path/to/output/dir> should exist New images from video will be named same as video name. As of now we generate only one image per video Output is json with key as video/image name and value as labels ``` ## Requirements - OpenCV with python support - Tika and Tensor flow running as per https://wiki.apache.org/tika/TikaAndVision#Step_1._Setup_REST_Server
