Aaron Edell

There are a few more boxes now; Tagbox: Is trainable now, so you can improve the model with your own images and have more customized image recognition. It can be improved on the fly, so it can be user generated as well. Videobox: Post videos to all these services and get back nice segmented results, including grouped unknown faces (so its easy to tag). Suggestionbox: Content...

What attracted me to Arlo originally was the wirelessness of it - but the router is annoying and not necessary. Also, there is a delay between when it detects motion and when it starts recording, so unless you position your cameras perfectly, you can totally miss whatever it was that triggered the camera.

After our car got stolen out of our driveway, I've been really interested in home security cams. But one thing I'd love to see is some machine learning applied to the videos stored in the cloud. I want to be alerted if a human face is detected somewhere (especially outside), not every time a car drives by. Try integrating something like Facebox (https://machinebox.io/docs/facebox) into the...

This is awesome. If you could add support for an icon for Medium that would be great. And yes - I think an e-mail signature is where I'd most likely want this page to go. Not the link to this page - but the page itself (obviously in much smaller form). I'm not sure where else I'd put this or send to people. I suppose it could be a link on my b-card or in my email. I'll have to experiment.

This is really cool. I've just launched a startup so I could see myself using this to meet potential customers and users in-person at events and tradeshows, especially since our target customers are developers and tech-oriented product people who are more likely to be on this network. Speaking of which - if you have a need to wrangle profile photos with face recognition, or moderate...

This is really cool! How are you accomplishing the summarization? Are you using something like Textbox to extract keyword frequency or something? https://machinebox.io/docs/textbox

This is super super super cool. You could use machine learning / visual search to let users upload an image of their favorite kind of architecture, and have that search for similar looking places. You could build it really easily with something like https://github.com/machinebox/visualsearch

It would be really cool if you could use something like Textbox (https://machinebox.io/docs/textbox) or Microsoft's text analysis API to do some natural language processing and auto-categorize the tasks, or make them searchable. You could also use it to auto-create reminders synced to a specific day or time like "Finish this by noon today" could turn into an actual time/date. Machine learning ftw!

This is really cool. You can also build your own version of this and customize it to maybe track your brand or something: https://github.com/machinebox/mood

This is really cool. Pair this up with something like Textbox from http://machinebox.io and you've got yourself a powerful ML powered experience in, like, an hour.