HyperAI
HyperAI초신경
홈
플랫폼
문서
뉴스
연구 논문
튜토리얼
데이터셋
백과사전
SOTA
LLM 모델
GPU 랭킹
컨퍼런스
전체 검색
소개
서비스 약관
개인정보 처리방침
한국어
HyperAI
HyperAI초신경
Toggle Sidebar
전체 사이트 검색...
⌘
K
Command Palette
Search for a command to run...
플랫폼
홈
SOTA
3D 다중 인물 자세 추정
3D Multi Person Pose Estimation On Shelf
3D Multi Person Pose Estimation On Shelf
평가 지표
PCP3D
평가 결과
이 벤치마크에서 각 모델의 성능 결과
Columns
모델 이름
PCP3D
Paper Title
TesseTrack (paper)
98.2
TesseTrack: End-to-End Learnable Multi-Person Articulated 3D Pose Tracking
QuickPose
98.1
QuickPose: Real-time Multi-view Multi-person Pose Estimation in Crowded Scenes
Zhang et al.
98.1
Lightweight Multi-person Total Motion Capture Using Sparse Multi-view Cameras
TesseTrack (correct)
97.9
TesseTrack: End-to-End Learnable Multi-Person Articulated 3D Pose Tracking
PlaneSweepPose
97.9
Multi-View Multi-Person 3D Pose Estimation with Plane Sweep Stereo
PRGN
97.7
Graph-Based 3D Multi-Person Pose Estimation Using Multi-View Images
UTTM
97.7
A Unified Multi-view Multi-person Tracking Framework
Faster VoxelPose
97.6
Faster VoxelPose: Real-time 3D Human Pose Estimation by Orthographic Projection
4D Association Pose
97.6
4D Association Graph for Realtime Multi-person Motion Capture Using Multiple Video Cameras
MvP
97.4
Direct Multi-view Multi-person 3D Pose Estimation
SmartEdgeSensor
97.4
Real-Time Multi-View 3D Human Pose Estimation using Semantic Feedback to Smart Edge Sensors
Part-aware Pose
97.39
Part-Aware Measurement for Robust Multi-View Multi-Human 3D Pose Estimation and Tracking
VTP
97.3
VTP: Volumetric Transformer for Multi-view Multi-person 3D Pose Estimation
VoxelTrack
97.1
VoxelTrack: Multi-Person 3D Human Pose Estimation and Tracking in the Wild
VoxelPose
97
VoxelPose: Towards Multi-Camera 3D Human Pose Estimation in Wild Environment
MVPose
96.9
Fast and Robust Multi-Person 3D Pose Estimation from Multiple Views
Multi Person Pose
96.9
Shape-aware Multi-Person Pose Estimation from Multi-View Images
crossview_3d_pose_tracking
96.8
Cross-View Tracking for Multi-Human 3D Pose Estimation at over 100 FPS
Cross-View
96.8
Cross-View Tracking for Multi-Human 3D Pose Estimation at over 100 FPS
Light3DPose
89.8
Light3DPose: Real-time Multi-Person 3D PoseEstimation from Multiple Views
0 of 24 row(s) selected.
Previous
Next
3D Multi Person Pose Estimation On Shelf | SOTA | HyperAI초신경