Applying AI to climbing … The Long Road to the 2020 Tokyo Olympics

RE: Applying AI to climbing … The Long Road to the 2020 Tokyo Olympics

albertabeef — Fri, 16 Apr 2021 16:50:17 GMT

Check out my new blog on the subject :
Applying AI to climbing … Deep Learning meets the "odd human" dataset

RE: Applying AI to climbing … The Long Road to the 2020 Tokyo Olympics

mbrown — Fri, 16 Apr 2021 15:58:23 GMT

I love the confluence your two passions in this blog! Can't wait to watch you 'send it' with the technical challenge!

/Matt

RE: Applying AI to climbing … The Long Road to the 2020 Tokyo Olympics

albertabeef — Wed, 14 Apr 2021 13:11:14 GMT

Just heard on the radio this morning that Japan may be entering a fourth wave, making the organization of the Olympics very challenging.
I do hope the event will take place and the athletes will be save and able to concentrate on their sport.
I heard that they will be monitored regularly for covid-19, and if they catch it, they are out.

Earning a 2020 Olympic medal, will not only represent being the best in your discipline, but also managing to not catch covid-19.

Crazy world we are now living in.

Mario.

RE: Applying AI to climbing … The Long Road to the 2020 Tokyo Olympics

neilk — Wed, 14 Apr 2021 08:54:30 GMT

Mario - fascinating project!

I rock climbed from the age of 16 through to my early 40s, but gave up because of a wife and children who were increasingly unhappy at some of the risks I was taking.

I briefly revisited the rocks in my mid 50s, with a professional guide. My ability to "see" the moves was only slightly diminished, but, sadly. I had lost my "bottle". I spent 3 very enjoyable days following on a tight rope!

I look forward to reading your next blog.

Neil

RE: Applying AI to climbing … The Long Road to the 2020 Tokyo Olympics

ralphjy — Mon, 12 Apr 2021 21:07:22 GMT

Hi Mario,

I'll be interested to see how well you do with this. My first thought is that you'll need a lot of image data to be able to discriminate the various moves. Is there an image database for rock climbing?

Maybe initially you could identify the route and based on other parameters (time, climber characteristics), you could identify which videos are of interest (and maybe which points are of interest) - and then try to focus on identifying a particular move at a particular section.

Almost seems like too hard a problem to solve with a 2D video image. It will be fun to see your "beta" for this problem [View:/resized-image/__size/16x16/__key/commentfiles/f7d226abd59f475c9d224a79e3f0ec07-61ed7963-373b-4ea7-8611-be1e75804572/contentimage_5F00_938.png:16:16].

-Ralph

RE: Applying AI to climbing … The Long Road to the 2020 Tokyo Olympics

DAB — Mon, 12 Apr 2021 19:54:15 GMT

Rock climbing would have been something to do back when I was in HS and at the peak of my physical capabilities.

Today I just watch and admire those who have the skill and passion to have fun.

DAB