3d 舞蹈同步

看起来很强大

unity驱动bvh跳舞：

脚飘动问题：

bvh和播放关节对应关系 zxy格式

bvh和播放关节对应关系 zyx的对应关系：

bvh播放器：

看起来很强大

GitHub - FORTH-ModelBasedTracker/MocapNET: We present MocapNET, a real-time method that estimates the 3D human pose directly in the popular Bio Vision Hierarchy (BVH) format, given estimations of the 2D body joints originating from monocular color images. Our contributions include: (a) A novel and compact 2D pose NSRM representation. (b) A human body orientation classifier and an ensemble of orientation-tuned neural networks that regress the 3D human pose by also allowing for the decomposition of the body to an upper and lower kinematic hierarchy. This permits the recovery of the human pose even in the case of significant occlusions. (c) An efficient Inverse Kinematics solver that refines the neural-network-based solution providing 3D human pose estimations that are consistent with the limb sizes of a target person (if known). All the above yield a 33% accuracy improvement on the Human 3.6 Million (H3.6M) dataset compared to the baseline method (MocapNET) while maintaining real-time performance

unity驱动bvh跳舞：

本机：My project (1)

脚飘动问题：

这句代码注释，脚就会飘动

anim.GetBoneTransform(HumanBodyBones.Hips).position = bvhPos[bp.root.name]*scaleRatio;

bvh和播放关节对应关系 zxy格式

Hips Hips
LeftUpLeg LeftUpperLeg
LeftLeg LeftLowerLeg
LeftFoot LeftFoot
RightUpLeg RightUpperLeg
RightLeg RightLowerLeg
RightFoot RightFoot
Spine Spine
Spine1 Chest
Neck Neck
LeftShoulder LeftShoulder
LeftArm LeftUpperArm
LeftForeArm LeftLowerArm
LeftHand LeftHand
RightShoulder RightShoulder
RightArm RightUpperArm
RightForeArm RightLowerArm
RightHand RightHand

bvh和播放关节对应关系 zyx的对应关系：

Hips Hips
LeftUpLeg RightUpperLeg
LeftLeg RightLowerLeg
LeftFoot RightFoot
RightUpLeg LeftUpperLeg
RightLeg LeftLowerLeg
RightFoot LeftFoot
Spine Spine
Spine1 Chest
Neck Neck
LeftShoulder RightShoulder
LeftArm RightUpperArm
LeftForeArm RightLowerArm
LeftHand RightHand
RightShoulder LeftShoulder
RightArm LeftUpperArm
RightForeArm LeftLowerArm
RightHand LeftHand