Introduces OR-Action benchmark for multi-role fine-grained actions in OR videos and a vision-only temporal model with multi-to-single view alignment that outperforms graph-based approaches.
Mvor: A multi-view rgb-d operating room dataset for 2D and 3D human pose estimation
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
cs.CV 3verdicts
UNVERDICTED 3representative citing papers
Ranking metrics AP and FPR-95 can be made perfect via Sinkhorn normalization even when assignment is already correct, while optimal ranking can still produce incorrect assignments.
LiCamPose combines multi-view RGB and LiDAR inputs via volumetric fusion, pretrains on synthetic data, and applies unsupervised adaptation to achieve robust single-frame 3D human pose estimation on multiple datasets.
citing papers explorer
-
OR-Action: Multi-Role Video Understanding with Fine-Grained Actions
Introduces OR-Action benchmark for multi-role fine-grained actions in OR videos and a vision-only temporal model with multi-to-single view alignment that outperforms graph-based approaches.
-
Ranking vs. Assignment: The Metric Mismatch in Multi-View Object Association
Ranking metrics AP and FPR-95 can be made perfect via Sinkhorn normalization even when assignment is already correct, while optimal ranking can still produce incorrect assignments.
-
LiCamPose: Combining Multi-View LiDAR and RGB Cameras for Robust Single-timestamp 3D Human Pose Estimation
LiCamPose combines multi-view RGB and LiDAR inputs via volumetric fusion, pretrains on synthetic data, and applies unsupervised adaptation to achieve robust single-frame 3D human pose estimation on multiple datasets.