2 comments

  • icelancer 1419 days ago
    This is against the OpenPose license, which strictly disallows the use of it for sports if anyone uses it for commercial purposes. I'd take this down or at least note it. They're known for being extremely litigious. It's not clear why this carveout exists but it's likely they licensed it to an NBA team or a provider for basketball (not going to get into how I know this).

    From their non-commercial licensing page:

    "The non-exclusive commercial license cannot be used in the field of Sports. ("Sports" shall mean any and all athletic competition between individuals, groups of individuals or teams.)"

    https://cmu.flintbox.com/#technologies/b820c21d-8443-4aa2-a4...

    • callmetonypls 1419 days ago
      Hi, thanks for telling me this. I'm definitely not using it for commercial purpose. I did read their whole license on their GitHub page.

      https://github.com/CMU-Perceptual-Computing-Lab/openpose/blo...

      I'm not really good at English, can you please help me with it?

      From their GitHub LICENSE:

      1. ACADEMIC OR NON-PROFIT ORGANIZATION NONCOMMERCIAL RESEARCH USE ONLY

      2. Licensor retains exclusive ownership of any copy of the Software (as defined below) licensed under this Agreement and hereby grants to Licensee a personal, non-exclusive, non-transferable license to use the Software for noncommercial research purposes, without the right to sublicense, pursuant to the terms and conditions of this Agreement.

      3. PERMITTED USES: The Software may be used for your own noncommercial internal research purposes.

      Here are my points:

      1. I build this only for fun, and I'm not making any money from it, which makes it non-commercial?

      2. I have mentioned that I implemented OpenPose and link to their GitHub repo.

      3. I have copy and paste all of their license into my repo license.

      4. Basically, I have no intention of stealing OpenPose and claim it as my own research. I just found out a cool implementation of their project. If anyone is interested in the pose analysis, I have provided the link on my README which will redirect him to the CMU's OpenPose repo.

      Is it still possible to have any license problem? I'm seriously not sure about it.

    • craftinator 1419 days ago
      It's definitely riding the edge of their license, but I don't see any content indicating "competition". While this could be used for an athletic competition, the author is only using it for personal, non competitive use on his GitHub. Best to be careful, make it clear it's not being used in competitive sports.
      • callmetonypls 1419 days ago
        Sure, thanks for checking it for me. I would add some extra information to my README. Telling that this project is only for noncommercial research use only.

        It's kinda weird that the LICENSE in CMU's OpenPose repo didn't indicate anything about sport or competition.

      • icelancer 1419 days ago
        I'm just saying through experience, the OpenPose people are very litigious and ornery. I think the project is a good one. But I also know what lies on the other side of the line.
  • callmetonypls 1420 days ago
    Hi guys, I know some of you may have already seen this project. I have made some big updates and add some interesting new features. I wanna share it with you guys and once again ask for more feedback.

    The original feature of this project is only shot analysis, it could trace the basketball shot and determine whether it's a score or miss. And now, it has a brand new feature! It's capable of analyzing the shooting pose. Besides displaying the angle of elbow and knee on every frame, it will also show the average pose analysis of each shot attempt. Release angle and release time are also included!

    1. I implemented OpenPose to analyze the shooting pose. The angle of elbow and knee is calculated with the data collected by OpenPose.

    2. The original shot analysis has a big problem which is that it would misdetect a head as a ball, and that will really mess up my algorithm. I solve this problem by avoiding the detections near the head coordinate collected from OpenPose.

    3. The shot release angle is calculated with the first two detection points of the shot analysis.

    4. Please note that there will be a relatively big error for the release time since it was calculated as the total time when the ball is in hand for each shot. I would be glad if anyone has a suggestion to solve this problem.

    Please feel free to check out the README on GitHub page for more detail.

    BTW, I would like to add more and more features. Please feel free to suggest what more analytics you would like to see on this project! I'm currently doing research on YOLOv4. I think it will significantly improve the efficiency of this project.

    • aaronbrethorst 1419 days ago
      Neat! To be frank, I don’t care about basketball in the slightest, but I do want to be able to use this to analyze my yoga poses and make suggestions about how I can improve my form.