Hierarchical Surface Prediction for 3D Object Reconstruction

Häne, Christian; Tulsiani, Shubham; Malik, Jitendra

doi:10.1109/3dv.2017.00054

Cited by 267 publications

(212 citation statements)

References 41 publications

Supporting

Mentioning

197

Contrasting

Order By: Relevance

“…We stress that unlike the baseline method, we did not re-train our model to handle the multi-view task. 256 3 grid resolution Tab.3 presents a comparison with the literature methods, conducted on the data provided by Hänee et al [13]. In order to compare with previous work which reported results in a grid resolution of 32 3 , pooling with stride 8 was applied to the predicted voxel grid generated at test time.…”

Section: Quantitative Resultsmentioning

confidence: 99%

Deep Meta Functionals for Shape Representation

Littwin

Wolf

2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

We present a new method for 3D shape reconstruction from a single image, in which a deep neural network directly maps an image to a vector of network weights. The network parametrized by these weights represents a 3D shape by classifying every point in the volume as either within or outside the shape. The new representation has virtually unlimited capacity and resolution, and can have an arbitrary topology. Our experiments show that it leads to more accurate shape inference from a 2D projection than the existing methods, including voxel-, silhouette-, and mesh-based methods. The code is available at: https://github. com/gidilittwin/Deep-Meta.

show abstract

Section: Quantitative Resultsmentioning

confidence: 99%

Deep Meta Functionals for Shape Representation

Littwin

Wolf

2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

show abstract

“…Most commonly, output 3D shapes are represented as voxel grids [4]. Using octrees instead of dense voxel grids [7], [18] allows to generate shapes of higher resolution. Multiple works concentrated on networks for predicting point clouds [6], [19] or meshes [20], [21].…”

Section: Related Workmentioning

confidence: 99%

Self-supervised 3D Shape and Viewpoint Estimation from Single Images for Robotics

Mees

Tatarchenko

Brox

et al. 2019

2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

We present a convolutional neural network for joint 3D shape prediction and viewpoint estimation from a single input image. During training, our network gets the learning signal from a silhouette of an object in the input image -a form of self-supervision. It does not require ground truth data for 3D shapes and the viewpoints. Because it relies on such a weak form of supervision, our approach can easily be applied to real-world data. We demonstrate that our method produces reasonable qualitative and quantitative results on natural images for both shape estimation and viewpoint prediction. Unlike previous approaches, our method does not require multiple views of the same object instance in the dataset, which significantly expands the applicability in practical robotics scenarios. We showcase it by using the hallucinated shapes to improve the performance on the task of grasping real-world objects both in simulation and with a PR2 robot.

show abstract

“…However, their main drawback is a low resolution because of prohibitively high computational cost from predicting every voxel in a 3D space. A recent improvement uses a hierarchical volumetric model that attempts to predict only those voxels at the surface of an object . This allows for significantly finer resolutions.…”

Section: D Computer Vision With Deep Learningmentioning

confidence: 99%

“…A recent improvement uses a hierarchical volumetric model that attempts to predict only those voxels at the surface of an object. 51 This allows for significantly finer resolutions.…”

Section: Depth Representationmentioning

confidence: 99%

3D computer vision based on machine learning with deep neural networks: A review

Bhowmik

2017

J Soc Info Display

View full text Add to dashboard Cite

Recent advances in the field of computer vision can be attributed to the emergence of deep learning techniques, in particular convolutional neural networks. Neural networks, partially inspired by the brain's visual cortex, enable a computer to “learn” the most important features of the images it is shown in relation to a specific, specified task. Given sufficient data and time, (deep) convolutional neural networks offer more easily designed, more generalizable, and significantly more accurate end‐to‐end systems than is possible with previously employed computer vision techniques. This review paper seeks to provide an overview of deep learning in the field of computer vision with an emphasis on recent progress in tasks involving 3D visual data. Through a backdrop of the mammalian visual processing system, we hope to also provide inspiration for future advances in automated visual processing.

show abstract

Hierarchical Surface Prediction for 3D Object Reconstruction

Cited by 267 publications

References 41 publications

Deep Meta Functionals for Shape Representation

Deep Meta Functionals for Shape Representation

Self-supervised 3D Shape and Viewpoint Estimation from Single Images for Robotics

3D computer vision based on machine learning with deep neural networks: A review

Contact Info

Product

Resources

About