在背身、跳跃等场景，骨骼识别不准确造成musepose没法使用，是否可以拆分单独的骨骼输入节点，这样更加灵活。例如骨骼这里有更好的方案https://github.com/AInseven/ComfyUI-fastblend
In scenarios such as back-to-back and jumping, inaccurate bone recognition makes Musepose unusable. Can we split separate bone input nodes for greater flexibility. For example, there is a better plan for bones here https://github.com/AInseven/ComfyUI-fastblend
在背身、跳跃等场景，骨骼识别不准确造成musepose没法使用，是否可以拆分单独的骨骼输入节点，这样更加灵活。例如骨骼这里有更好的方案https://github.com/AInseven/ComfyUI-fastblend In scenarios such as back-to-back and jumping, inaccurate bone recognition makes Musepose unusable. Can we split separate bone input nodes for greater flexibility. For example, there is a better plan for bones here https://github.com/AInseven/ComfyUI-fastblend