Are you still working on this? This is a problem I have been thinking about since a while now (I am a beginner when it comes to climbing as well) I think estimating the ideal body positions (centre of gravity, feet and hands) through the boulder problem progression would be so cool and handy. This is the coolest thing I saw today, AWESOME. Keep climbing!

I prompted VLM Run’s visual agent Orion to segment all of the blue bouldering holds, and it did a good job! It is interesting that now we can prompt VLMs to segment all of the holds, rather than creating a new dataset from scratch to train a model.

With holds detection + pose estimation, I can show how each hold gets activated as a hand or foot uses it. Once we touch the final hold with both hands, the route is completed, and I show the overall path of my torso midpoint.

A tool like this could help climbers understand their movement better. I’m still very much a beginner at bouldering, so it would be great to get quantitative feedback.

There are definitely things to improve, but overall I’m encouraged by this first demo.

Models used:

- VLM Run’s Orion for segmentation (https://www.vlm.run/)

- ViTPose+ Huge for pose estimation (via Hugging Face)

- RT-DETR for person detection (via Hugging Face)

Shoutout to Daniel Reiff and his bouldering + computer vision project for the inspiration!

Link to Daniel Reiff's bouldering + computer vision blog: https://blog.roboflow.com/bouldering/