HyperAI
HyperAI超神经
首页
算力平台
文档
资讯
论文
教程
数据集
百科
SOTA
LLM 模型天梯
GPU 天梯
顶会
开源项目
全站搜索
关于
中文
HyperAI
HyperAI超神经
Toggle sidebar
全站搜索…
⌘
K
全站搜索…
⌘
K
首页
SOTA
三维人体姿态估计
3D Human Pose Estimation On Mpi Inf 3Dhp
3D Human Pose Estimation On Mpi Inf 3Dhp
评估指标
MPJPE
PCK
评测结果
各个模型在此基准测试上的表现结果
Columns
模型名称
MPJPE
PCK
Paper Title
Repository
VNect (Augm.)
124.7
76.6
VNect: Real-time 3D Human Pose Estimation with a Single RGB Camera
HMR
124.2
72.9
End-to-end Recovery of Human Shape and Pose
Single-Shot Multi-Person
122.2
75.2
Single-Shot Multi-Person 3D Pose Estimation From Monocular RGB
Mehta
117.6
75.7
Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervision
-
PONet
115.0
76.1
PONet: Robust 3D Human Pose Estimation via Learning Orientations Only
-
DenseRaC
114.2
76.9
DenseRaC: Joint 3D Pose and Shape Estimation by Dense Render-and-Compare
-
Pose Consensus (monocular)
112.1
80.6
Consensus-based Optimization for 3D Human Pose Estimation in Camera Coordinates
GeoRep (fully-supervised)
110.8
80.2
Weakly-Supervised 3D Human Pose Learning via Multi-view Images in the Wild
-
XFormer (HRNet)
109.8
-
XFormer: Fast and Accurate Monocular 3D Body Capture
-
EpipolarPose (fully-supervised)
108.99
77.5
Self-Supervised Learning of 3D Human Pose using Multi-view Geometry
SPIN
105.2
76.4
Learning to Reconstruct 3D Human Pose and Shape via Model-fitting in the Loop
DSR
104.7
-
Learning to Regress Bodies from Images using Differentiable Semantic Rendering
CanonPose
104
77
CanonPose: Self-Supervised Monocular 3D Human Pose Estimation in the Wild
RSC-Net
103.36
-
3D Human Shape and Pose from a Single Low-Resolution Image with Self-Supervised Learning
PoseNet3D
102.4
81.9
PoseNet3D: Learning Temporally Consistent 3D Human Pose via Knowledge Distillation
DynaBOA
101.5
79.5
Out-of-Domain Human Mesh Reconstruction via Dynamic Bilevel Online Adaptation
EvoSkeleton
99.7
81.2
Cascaded deep monocular 3D human pose estimation with evolutionary training data
XNect (SelecSLS)
98.4
82.8
XNect: Real-time Multi-Person 3D Motion Capture with a Single RGB Camera
DDT
97.8
-
DiffMesh: A Motion-aware Diffusion Framework for Human Mesh Recovery from Videos
-
RepNet (3DHP)
97.8
82.5
RepNet: Weakly Supervised Training of an Adversarial Reprojection Network for 3D Human Pose Estimation
0 of 116 row(s) selected.
Previous
Next
3D Human Pose Estimation On Mpi Inf 3Dhp | SOTA | HyperAI超神经