Neural Correspondence Field for Object Pose Estimation
원문 초록
We propose a method for estimating the 6DoF pose of a rigid object with an available 3D model from a single RGB image. Unlike classical correspondence-based methods which predict 3D object coordinates at pixels of the input image, the proposed method predicts 3D object coordinates at 3D query points sampled in the camera frustum. The move from pixels to 3D points, which is inspired by recent PIFu-style methods for 3D reconstruction, enables reasoning about the whole object, including its (self-)occluded parts. For a 3D query point associated with a pixel-aligned image feature, we train a fully-connected neural network to predict: (i) the corresponding 3D object coordinates, and (ii) the signed distance to the object surface, with the first defined only for query points in the surface vicinity. We call the mapping realized by this network as Neural Correspondence Field. The object pose is then robustly estimated from the predicted 3D-3D correspondences by the Kabsch-RANSAC algorithm. The proposed method achieves state-of-the-art results on three BOP datasets and is shown superior especially in challenging cases with occlusion. The project website is at: linhuang17.github.io/NCF.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
SpatialTrackerV2: 3D Point Tracking Made Easy
Consensus-Driven Uncertainty for Robotic Grasping based on RGB Perception
RGBTrack: Fast, Robust Depth-Free 6D Pose Estimation and Tracking
AutoPartGen: Autogressive 3D Part Generation and Discovery
BRUM: Robust 3D Vehicle Reconstruction from 360 Sparse Images
DreamGrasp: Zero-Shot 3D Multi-Object Reconstruction from Partial-View Images for Robotic Manipulation
토론
이 논문과 연결된 포럼 글이 아직 없습니다.