computer-vision-3

Single-View 3D Reconstruction

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in single-view-3d-reconstruction-3

Trend

Dataset

Best Model

Actions

ShapeNet

ShapeNetCore

Common Objects in 3D

Libraries

i

Use these libraries to find single-view-3d-reconstruction-3 models and implementations

Datasets

Subtasks

3D Semantic Scene Completion from a single RGB image

Most implemented papers

Learning Implicit Fields for Generative Shape Modeling

Zhiqin Chen, Hao Zhang•Wed Dec 05 2018

By replacing conventional decoders by the implicit decoder for representation learning and shape generation, this work demonstrates superior results for tasks such as generative shape modeling, interpolation, and single-view 3D reconstruction, particularly in terms of visual quality.

1769

Content

Common Objects in 3D

TransProteus

Market-HQ

ATR

CUB-200-2011

ABO

0

PQ-NET: A Generative Part Seq2Seq Network for 3D Shapes

Baoquan Chen, Rundi Wu, Kai Xu, Yixin Zhuang•Sun Nov 24 2019

PQ-NET is introduced, a deep neural network which represents and generates 3D shapes via sequential part assembly which encodes a sequence of part features into a latent vector of fixed size and reconstructs the 3D shape, one part at a time, resulting in a sequential assembly.

186 0

Paper Graph

DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstruction

Duygu Ceylan, R. Měch, U. Neumann, Qiangeng Xu, Weiyue Wang•Sat May 25 2019

DISN, a Deep Implicit Surface Network which can generate a high-quality detail-rich 3D mesh from an 2D image by predicting the underlying signed distance fields by combining global and local features, achieves the state-of-the-art single-view reconstruction performance.

600 0

Paper Graph

Learning to Detect 3D Reflection Symmetry for Single-View Reconstruction

Yichao Zhou, Shichen Liu, Yi Ma•Tue Jun 16 2020

This work presents a geometry-based end-to-end deep learning framework that first detects the mirror plane of reflection symmetry that commonly exists in man-made objects and then predicts depth maps by finding the intra-image pixel-wise correspondence of the symmetry.

7 0

Paper Graph

Soft Rasterizer: A Differentiable Renderer for Image-Based 3D Reasoning

Tianye Li, Weikai Chen, Shichen Liu•Tue Apr 02 2019

This work proposes a truly differentiable rendering framework that is able to directly render colorized mesh using differentiable functions and back-propagate efficient supervision signals to mesh vertices and their attributes from various forms of image representations, including silhouette, shading and color images.

766 0

Paper Graph

3D-LMNet: Latent Embedding Matching for Accurate and Diverse 3D Point Cloud Reconstruction from a Single Image

Mayank Agarwal, R. Venkatesh Babu, Priyanka Mandikal, L. NavaneetK.•Thu Jul 19 2018

3D-LMNet, a latent embedding matching approach for 3D reconstruction, is proposed, which outperform state-of-the-art approaches on the task of single-view3D reconstruction on both real and synthetic datasets while generating multiple plausible reconstructions, demonstrating the generalizability and utility of the approach.

170 0

Paper Graph

Learning Single-View 3D Reconstruction with Limited Pose Supervision

Yin Cui, Serge J. Belongie, Bharath Hariharan, Guandao Yang•Fri Sep 07 2018

A unified framework is presented that can combine both types of supervision: a small amount of camera pose annotations are used to enforce pose-invariance and view-point consistency, and unlabeled images combined with an adversarial loss are use to enforce the realism of rendered, generated models.

59 0

Paper Graph

Domain-Adaptive Single-View 3D Reconstruction

Negar Rostamzadeh, Sungjin Ahn, Pedro H. O. Pinheiro•Mon Dec 03 2018

This paper imposes domain confusion between natural and synthetic image representations to reduce the distribution gap, and imposes the reconstruction to be `realistic' by forcing it to lie on a (learned) manifold of realistic object shapes.

27 0

Paper Graph

Learning to Predict 3D Objects with an Interpolation-based Differentiable Renderer

S. Fidler, J. Lehtinen, Huan Ling, Alec Jacobson, Edward James Smith, Wenzheng Chen, Jun Gao•Fri Aug 02 2019

A differentiable rendering framework which allows gradients to be analytically computed for all pixels in an image and to view foreground rasterization as a weighted interpolation of local properties and background rasterized as a distance-based aggregation of global geometry.

378 0

Paper Graph

Hyperparameter-Free Losses for Model-Based Monocular Reconstruction

Xavier Giró-i-Nieto, Eduard Ramon, Guillermo Ruiz, Thomas Batard•Thu Aug 15 2019

This work proposes novel hyperparameter-free losses for single view 3D reconstruction with morphable models (3DMM) and proposes a novel implicit regularization technique based on random virtual projections that does not require additional 2D or 3D annotations.

1 0

Paper Graph

Adding a benchmark result helps the community track progress.