Cool project but please delete it. You’re taking the fun part out from bouldering and just leaving us the labour ;) Edit: apprently I wasnt the first to make this joke. I emphasise that this is Really a cool project. Great job!
Show HN: AI climbing coach – visualize how to climb any route based on your body
131–140 of 149 posts
Re: Show HN: AI climbing coach – visualize how to climb any route based on your body
#132A few thoughts as I've done academic research & built products in this area: - if you're using SMPL body parameters this will have to stay research / open-source - is this leveraging some sort of monocular depth estimation to estimate the wall in 3D space? Also, do you have assumed camera parameters, or is that also estimated? If there isn't any depth information, this will be highly inaccurate on any cliff routes, b…
Yes, this will be open source. This is an end-end system that just takes in video frames. Camera parameters are one of the things that is predicted. It gives promising results for a wide variety of environments (cliffs, diff types of bouldering walls, diff outdoor walls, etc.), though not always accurate. Path planning is also part of the end-end system. Will share more details in the paper.
sweet can't wait to read it! I'll also be at CVPR this year if you're presenting
Re: Show HN: AI climbing coach – visualize how to climb any route based on your body
#133Very cool project. I will try this out when you release it
Re: Show HN: AI climbing coach – visualize how to climb any route based on your body
#134Earlier quoted context omitted.
It's 2024 and some people are gonna gatekeep that if you're not shoving cams in some bigwall granite, it's not climbing?
This one-upmanship is very much part of climbing culture. From the top down, the hierarchy is: free soloists (climbing at height without any protection), trad climbers (carry the rope up with them, set their own protection), sports lead climbers (carry the rope up with them, bolts are set into the wall), top rope climbers (rope already there dangling from the top). Right down the bottom, you have aid climbing, when y…
Aid climbing isn't respected on smaller or well-established cliffs, but most understand that it has a place. All the great routes started as aid routes, only being climbed "free" years later. Aid is also a very useful skill in the rain or rescue situations when friction disappears.
Re: Show HN: AI climbing coach – visualize how to climb any route based on your body
#135I feel like I’m the right audience for this - a not very good climber with unusual distribution of limbs. But I’m not interested - part of the fun in climbing is on sight climbs and solving problems yourself.
Man, where is the hacking spirit? OP posts a very cool, impressive hack and all everyone can talk about in the comments is how it actually isn't interesting to them.
If somebody solves the 'hacking' part for you, a lot of fun disappears. Yes they probably had tons of fun, but you lose some of it.
Figuring a route for the first time is by far the most rewarding part of climbing (although I love all of it). Same with kids - if they grok something on their own, what a reaction and reward compared to being told how to do it.
Re: Show HN: AI climbing coach – visualize how to climb any route based on your body
#136I feel like I’m the right audience for this - a not very good climber with unusual distribution of limbs. But I’m not interested - part of the fun in climbing is on sight climbs and solving problems yourself.
Man, where is the hacking spirit? OP posts a very cool, impressive hack and all everyone can talk about in the comments is how it actually isn't interesting to them.
It's cool as an academic exercise, but there are things in life where the struggle is kind of the point. The messy learning process, asking other humans for help, is better for your growth.
Re: Show HN: AI climbing coach – visualize how to climb any route based on your body
#137This is rad and I can't wait to try it out for myself
Re: Show HN: AI climbing coach – visualize how to climb any route based on your body
#138Very cool idea but I wonder how well this can possibly work in practice. So many bouldering problems hinge on small things, such as how a little crimp hold is oriented or how deep the groove in it is. It can be hard to determine this even in person until you get up close or try to grab it.
Additionally, this model seems to take into account the body type but not other relevant factors, such as climber's strength and flexibility/mobility. Even in the demo, that suggested move of putting the right foot high and pushing off the jug up and to the right could not be done by a lot of new climbers. This parallels what I've personally seen when watching others climb. For example, some climbers are exceptionally strong and can basically campus their way up a route. Watching their beta is irrelevant to me. I don't have the same strength to pull those moves.
Re: Show HN: AI climbing coach – visualize how to climb any route based on your body
#139I feel like I’m the right audience for this - a not very good climber with unusual distribution of limbs. But I’m not interested - part of the fun in climbing is on sight climbs and solving problems yourself.
As an experienced, very poor climber myself, I enjoy the process of problem solving. However, knowing many experienced, very good climbers - studying beta is how many of them excel. As an aside, the term beta also came from the use of beta tapes by climbers to record themselves climbing so they could study the minutiae of their movement to find improvements. In higher levels of bouldering especially, a nuanced and firm understanding of beta is everything - so careful analysis of your movement can help identify areas of improvement.
I believe this is consistent with most elite (or elite-aspiring) athletes from many sports.
While I personally enjoy the problem-solving aspect of climbing (when I rarely do get out), I absolutely see the value in this project (and other climbing apps that thrive on beta sharing)
Re: Show HN: AI climbing coach – visualize how to climb any route based on your body
#140This has always been a dream, because there is a "perfect" way for your body to climb a route but it would be insanely complex to calculate, especially when factoring in every tiny physiological aspect like ankle strength and hamstring flexibility.
This is a really cool step though.
Maybe a more useful variant with the current tech would be isolating a single move and seeing every variation the AI can think up.