6D Pose Estimation On Linemod

评估指标

Accuracy

Mean ADD

Mean IoU

评测结果

各个模型在此基准测试上的表现结果

				Paper Title	Repository
RNNPose	-	97.37	-	RNNPose: Recurrent 6-DoF Object Pose Refinement with Robust Correspondence Field Estimation and Pose Optimization
EfficientPose	-	97.35	-	EfficientPose: An efficient, accurate and scalable end-to-end 6D multi object pose estimation approach
YOLOv5-6D	-	96.84	-	YOLOv5-6D: Advancing 6-DoF Instrument Pose Estimation in Variable X-Ray Imaging Geometries	-
EPro-PnP-6DoF v2	-	96.36	-	EPro-PnP: Generalized End-to-End Probabilistic Perspective-n-Points for Monocular Object Pose Estimation
RePOSE	-	96.1	-	RePOSE: Fast 6D Object Pose Refinement via Deep Texture Rendering
EPro-PnP-6DoF v1	-	95.8	-	EPro-PnP: Generalized End-to-End Probabilistic Perspective-n-Points for Monocular Object Pose Estimation
ROPE	-	95.61	-	Occlusion-Robust Object Pose Estimation with Holistic Representation
DPOD	-	95.2	-	DPOD: 6D Pose Object Detector and Refiner
HRNet (DSNT + BPnP)	-	93.3	-	End-to-End Learnable Geometric Vision by Backpropagating PnP Optimization
HybridPose	-	91.3	-	HybridPose: 6D Object Pose Estimation under Hybrid Representations
CDPN	98.1%	89.86	-	CDPN: Coordinates-Based Disentangled Pose Network for Real-Time RGB-Based 6-DoF Object Pose Estimation	-
PoseCNN + DeepIM	97.5	88.6	-	DeepIM: Deep Iterative Matching for 6D Pose Estimation
E2E6DoF	-	86.8	-	End-to-End Differentiable 6DoF Object Pose Estimation with Local and Global Constraints	-
PVNet	99%	86.27	-	PVNet: Pixel-wise Voting Network for 6DoF Pose Estimation
CullNet	97.7%	78.3	-	CullNet: Calibrated and Pose Aware Confidence Scores for Object Pose Estimation
SSD-6D	-	76.3	99.4	SSD-6D: Making RGB-based 3D detection and 6D pose estimation great again
Keypoint Detector Localization	94.5%	72.6	-	Estimating 6D Pose From Localizing Designated Surface Keypoints
Single-shot Deep CNN	90.37%	55.95	99.92	Real-Time Seamless Single Shot 6D Object Pose Prediction
BB8	83.9%	43.6	-	BB8: A Scalable, Accurate, Robust to Partial Occlusion Method for Predicting the 3D Poses of Challenging Objects without Using Depth
Pix2Pose	-	32	-	Pix2Pose: Pixel-Wise Coordinate Regression of Objects for 6D Pose Estimation

0 of 22 row(s) selected.