Technology

19 readers

2 users here now

This magazine is dedicated to discussions on the latest developments, trends, and innovations in the world of technology. Whether you are a tech enthusiast, a developer, or simply curious about the latest gadgets and software, this is the place for you. Here you can share your knowledge, ask questions, and engage in discussions on topics such as artificial intelligence, robotics, cloud computing, cybersecurity, and more. From the impact of technology on society to the ethical considerations of new technologies, this category covers a wide range of topics related to technology. Join the conversation and let's explore the ever-evolving world of technology together!

founded 1 year ago

Is there an opensource package that uses a ML model to identify people/objects in video content? (kbin.social)

submitted 5 months ago by DarthYoshiBoy@kbin.social to c/tech@kbin.social

2 comments fedilink hide all child comments

I'm looking for something that I could run locally and turn loose on a collection of videos to get a quick list of tags for each piece of content.

e.g. A video of a cat playing in a front yard on a sunny day would generate a collection of tags like ["Cat", "Grass", "Sidewalk", "Sunny", "Flower", "Dirt"] or a video of children playing on a playground would generate ["Child", "Slide", "Swing", "Seesaw", "Kids"]

There seem to be a number of online products that will do this sort of thing for YouTube videos or allow you to upload content to their cloud for analysis (and often for a decent price) but I don't want to run everything through the internet as it seems like I'd spend more time uploading stuff than it'd be worth the bother.

It seems like OpenCV might be capable of doing something like this, but I haven't found anyone speaking of its use without having to first train your own model which would probably reduce the effectiveness of this approach as I'd have to go tag all my own content first to teach the model how to do it?

top 2 comments

sorted by: hot top controversial new old

[–] pearsaltchocolatebar@discuss.online 0 points 5 months ago (1 children)

I'm also interested in this. I want to set up a camera and speaker to automatically yell at my dog when he's digging up the garden beds.

[–] DarthYoshiBoy@kbin.social 1 points 5 months ago

OpenCV may be fairly set for purpose for your needs already. If you could train the Cascade classifier (See: https://docs.opencv.org/4.x/dc/d88/tutorial_traincascade.html) with enough data of what constitutes your dog in the garden, you'd be able to string a video feed through a script and have it trigger audio playback whenever it identifies the naughty doggo. (See: https://docs.opencv.org/4.x/db/d28/tutorial_cascade_classifier.html)