Gradient-SDF: A Semi-Implicit Surface Representation for 3D Reconstruction

Sommer, Christiane; Sang, Lu; Schubert, David; Cremers, Daniel

doi:10.1109/cvpr52688.2022.00618

Cited by 12 publications

(1 citation statement)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Canelhas et al utilize SDF to represent the alignment error and estimate the motion of the camera [ 17 ]. Sommer et al [ 18 ] proposed gradient-SDF to describe 3D geometry that combines the advantages of implicit and explicit representations. Grinvald et al [ 19 ] extended the TSDF model using information about object categories.…”

Section: Related Workmentioning

confidence: 99%

Method for Estimating Road Impulse Ahead of Vehicles in Urban Environment with Microelectromechanical System Three-Dimensional Sensor

Zhao,

Wang,

Wang

et al. 2024

Sensors

View full text Add to dashboard Cite

Most automated vehicles (AVs) are equipped with abundant sensors, which enable AVs to improve ride comfort by sensing road elevation, such as speed bumps. This paper proposes a method for estimating the road impulse features ahead of vehicles in urban environments with microelectromechanical system (MEMS) light detection and ranging (LiDAR). The proposed method deploys a real-time estimation of the vehicle pose to solve the problem of sparse sampling of the LiDAR. Considering the LiDAR error model, the proposed method builds the grid height measurement model by maximum likelihood estimation. Moreover, it incorporates height measurements with the LiDAR error model by the Kalman filter and introduces motion uncertainty to form an elevation weight method by confidence eclipse. In addition, a gate strategy based on the Mahalanobis distance is integrated to handle the sharp changes in elevation. The proposed method is tested in the urban environment. The results demonstrate the effectiveness of our method.

show abstract

Section: Related Workmentioning

confidence: 99%

Method for Estimating Road Impulse Ahead of Vehicles in Urban Environment with Microelectromechanical System Three-Dimensional Sensor

Zhao,

Wang,

Wang

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

Neural Vector Fields for Implicit Surface Representation and Inference

Mello Rella,

Chhatkuli,

Konukoglu

et al. 2024

Int J Comput Vis

View full text Add to dashboard Cite

Neural implicit fields have recently shown increasing success in representing, learning and analysis of 3D shapes. Signed distance fields and occupancy fields are still the preferred choice of implicit representations with well-studied properties, despite their restriction to closed surfaces. With neural networks, unsigned distance fields as well as several other variations and training principles have been proposed with the goal to represent all classes of shapes. In this paper, we develop a novel and yet a fundamental representation considering unit vectors in 3D space and call it Vector Field (VF). At each point in $$\mathbb {R}^3$$ R 3 , VF is directed to the closest point on the surface. We theoretically demonstrate that VF can be easily transformed to surface density by computing the flux density. Unlike other standard representations, VF directly encodes an important physical property of the surface, its normal. We further show the advantages of VF representation, in learning open, closed, or multi-layered surfaces. We show that, thanks to the continuity property of the neural optimization with VF, a separate distance field becomes unnecessary for extracting surfaces from the implicit field via Marching Cubes. We compare our method on several datasets including ShapeNet where the proposed new neural implicit field shows superior accuracy in representing any type of shape, outperforming other standard methods. Codes are available at https://github.com/edomel/ImplicitVF.

show abstract

Autonomous Manipulation Learning for Similar Deformable Objects via Only One Demonstration

Ren,

Chen,

Cong

2023

2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

Learning robust and generalizable manipulation skills from demonstrations remains a key challenge in robotics, with broad applications in industrial automation and service robotics. While recent imitation learning methods have achieved impressive results, they often require large amounts of demonstration data and struggle to generalize across different spatial variants. In this work, we present a novel framework that learns manipulation skills from as few as 10 demonstrations, yet still generalizes to spatial variants such as different initial object positions and camera viewpoints. Our framework consists of two key modules: Semantic Guided Perception (SGP), which constructs task-focused, spatially aware 3D point cloud representations from RGB-D inputs; and Spatial Generalized Decision (SGD), an efficient diffusion-based decision-making module that generates actions via denoising. To effectively learn generalization ability from limited data, we introduce a critical spatially equivariant training strategy that captures the spatial knowledge embedded in expert demonstrations. We validate our framework through extensive experiments on both simulation benchmarks and real-world robotic systems. Our method demonstrates a 60-70% improvement in success rates over state-of-the-art approaches on a series of challenging tasks, even with substantial variations in object poses and camera viewpoints. This work shows significant potential for advancing efficient, generalizable manipulation skill learning in real-world applications. 1

show abstract

Gradient-SDF: A Semi-Implicit Surface Representation for 3D Reconstruction

Cited by 12 publications

References 23 publications

Method for Estimating Road Impulse Ahead of Vehicles in Urban Environment with Microelectromechanical System Three-Dimensional Sensor

Method for Estimating Road Impulse Ahead of Vehicles in Urban Environment with Microelectromechanical System Three-Dimensional Sensor

Neural Vector Fields for Implicit Surface Representation and Inference

Autonomous Manipulation Learning for Similar Deformable Objects via Only One Demonstration

Contact Info

Product

Resources

About