← 3D Human Pose Estimation
Papers with Code · CC BY-SA 4.0

Human3.6M

아카이브 평가 88건 · 마지막 적재/변경

평가 기록

과거 수집 기록이며 최신 순위가 아닙니다. 데이터셋 버전·split·실험 조건과 지표 방향이 검증되지 않은 결과는 직접 비교할 수 없습니다. 날짜는 논문 날짜이며 최신 평가일을 뜻하지 않습니다.

표가 화면보다 넓으면 좌우로 스크롤하여 논문과 구현 링크를 확인하세요.

원본 기록 순서 · 순위 미부여
모델원본 지표·값논문구현
StridedTransformer (T=351)
Average MPJPE (mm)
43.7
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Exploiting Temporal Contexts with Strided Transformer for 3D Human Pose Estimation
2021-03-26
코드 1
StridedTransformer (T=243)
Average MPJPE (mm)
44
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Exploiting Temporal Contexts with Strided Transformer for 3D Human Pose Estimation
2021-03-26
코드 1
Anatomy3D
Average MPJPE (mm)
44.1
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Anatomy-aware 3D Human Pose Estimation with Bone-based Pose Decomposition
2020-02-24
코드 1
P-STMO-S (N=81)
Average MPJPE (mm)
44.1
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
P-STMO: Pre-Trained Spatial Temporal Many-to-One Model for 3D Human Pose Estimation
2022-03-15
코드 1
PoseFormer (f=81)
Average MPJPE (mm)
44.3
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
3D Human Pose Estimation with Spatial and Temporal Transformers
2021-03-18
코드 1 · 코드 2 · 코드 3
RIE (T=243 CPN)
Average MPJPE (mm)
44.3
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Improving Robustness and Accuracy via Relative Information Encoding in 3D Human Pose Estimation
2021-07-29
코드 1
Probabilistic Monocular (T=200)
Average MPJPE (mm)
44.3
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Probabilistic Monocular 3D Human Pose Estimation with Normalizing Flows
2021-07-29
코드 1
Shape-aware SMPL
Average MPJPE (mm)
44.4
Using 2D ground-truth joints
No
Multi-View or Monocular
Multi-View
Shape-Aware Human Pose and Shape Reconstruction Using Multi-View Images
2019-08-26
연결 없음
TesseTrack (Monocular)
Average MPJPE (mm)
44.6
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
TesseTrack: End-to-End Learnable Multi-Person Articulated 3D Pose Tracking
2021-06-16
연결 없음
SRNet (T=243)
Average MPJPE (mm)
44.8
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
SRNet: Improving Generalization in 3D Human Pose Estimation with a Split-and-Recombine Approach
2020-07-18
코드 1
Attention (T=243 CPN)
Average MPJPE (mm)
44.8
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Enhanced 3D Human Pose Estimation from Videos by using Attention-Based Neural Network with Dilated Convolutions
2021-03-04
연결 없음
Virtual Bones (T=243 CPN)
Average MPJPE (mm)
44.8
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Motion Projection Consistency Based 3D Human Pose Estimation with Virtual Bones from Monocular Videos
2021-06-28
연결 없음
Pose Consensus (multi-view, est. calib.)
Average MPJPE (mm)
45
Using 2D ground-truth joints
No
Multi-View or Monocular
Multi-View
Consensus-based Optimization for 3D Human Pose Estimation in Camera Coordinates
2019-11-21
코드 1
HEMlets Pose
Average MPJPE (mm)
45.1
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
HEMlets Pose: Learning Part-Centric Heatmap Triplets for Accurate 3D Human Pose Estimation
2019-10-26
연결 없음
Attention3DHumanPose (T=243 CPN)
Average MPJPE (mm)
45.1
Using 2D ground-truth joints
No
Multi-View or Monocular
Multi-View
Attention Mechanism Exploits Temporal Contexts: Real-Time 3D Human Pose Reconstruction
2020-06-01
코드 1
StridedTransformer (T=81)
Average MPJPE (mm)
45.4
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Exploiting Temporal Contexts with Strided Transformer for 3D Human Pose Estimation
2021-03-26
코드 1
DC-GCT(T=1)
Average MPJPE (mm)
46.1
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Double-chain Constraints for 3D Human Pose Estimation in Images and Videos
2023-08-10
코드 1
Trajectory Space Factorization (50 frames)
Average MPJPE (mm)
46.6
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Trajectory Space Factorization for Deep Video-Based 3D Human Pose Estimation
2019-08-22
코드 1
VideoPose3D (T=243)
Average MPJPE (mm)
46.8
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
PA-MPJPE
36.5
3D human pose estimation in video with temporal convolutions and semi-supervised training
2018-11-28
코드 1 · 코드 2 · 코드 3 · 코드 4 · 코드 5 · 코드 6 · 코드 7 · 코드 8 · 코드 9 · 코드 10
StridedTransformer (T=27)
Average MPJPE (mm)
46.9
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Exploiting Temporal Contexts with Strided Transformer for 3D Human Pose Estimation
2021-03-26
코드 1
Virtual Bones (T=9 CPN)
Average MPJPE (mm)
47.4
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Motion Projection Consistency Based 3D Human Pose Estimation with Virtual Bones from Monocular Videos
2021-06-28
연결 없음
Skeletal GNN
Average MPJPE (mm)
47.9
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Learning Skeletal Graph Neural Networks for Hard 3D Pose Estimation
2021-08-16
연결 없음
GraphMLP
Average MPJPE (mm)
48
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
GraphMLP: A Graph MLP-Like Architecture for 3D Human Pose Estimation
2022-06-13
코드 1
STRGCN (T=7)
Average MPJPE (mm)
48.8
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Exploiting Spatial-Temporal Relationships for 3D Pose Estimation via Graph Convolutional Networks
2019-10-01
코드 1
STRGCN (T=3)
Average MPJPE (mm)
49.1
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Exploiting Spatial-Temporal Relationships for 3D Pose Estimation via Graph Convolutional Networks
2019-10-01
코드 1
DiffPyramid (CPN)
Average MPJPE (mm)
49.2
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2025-06-03
코드 1
Dual network
Average MPJPE (mm)
49.31
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Dual networks based 3D Multi-Person Pose Estimation from Monocular Video
2022-05-02
코드 1
Modulated-GCN
Average MPJPE (mm)
49.4
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Modulated Graph Convolutional Network for 3D Human Pose Estimation
2021-01-01
코드 1
MTF-Transformer (M=0.4, T=7, N=1)
Average MPJPE (mm)
49.4
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Adaptive Multi-view and Temporal Fusing Transformer for 3D Human Pose Estimation
2021-10-11
연결 없음
PGFormer (CPN)
Average MPJPE (mm)
49.5
Using 2D ground-truth joints
No
Multi-View or Monocular
Monocular
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2025-06-03
코드 1