3D Object Reconstruction From A Single Image

9 papers with code • 2 benchmarks • 2 datasets

Image: Fan et al

Most implemented papers

Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM

lh3/bwa 16 Mar 2013

Summary: BWA-MEM is a new alignment algorithm for aligning sequence reads or long query sequences against a large reference genome such as human.

A Point Set Generation Network for 3D Object Reconstruction from a Single Image

fanhqme/PointSetGeneration CVPR 2017

Our final solution is a conditional shape sampler, capable of predicting multiple plausible 3D point clouds from an input image.

PIFuHD: Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization

facebookresearch/pifuhd CVPR 2020

Although current approaches have demonstrated the potential in real world settings, they still fail to produce reconstructions with the level of detail often present in the input images.

PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization

shunsukesaito/PIFu ICCV 2019

We introduce Pixel-aligned Implicit Function (PIFu), a highly effective implicit representation that locally aligns pixels of 2D images with the global context of their corresponding 3D object.

Occlusion-Net: 2D/3D Occluded Keypoint Localization Using Graph Networks

dineshreddy91/Occlusion_Net CVPR 2019

Central to this work is a trifocal tensor loss that provides indirect self-supervision for occluded keypoint locations that are visible in other views of the object.

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

mees/self-supervised-3D 17 Oct 2019

We present a convolutional neural network for joint 3D shape prediction and viewpoint estimation from a single input image.

From Image Collections to Point Clouds with Self-supervised Shape and Pose Networks

val-iisc/ssl_3d_recon CVPR 2020

We learn both 3D point cloud reconstruction and pose estimation networks in a self-supervised manner, making use of differentiable point cloud renderer to train with 2D supervision.

SDF-SRN: Learning Signed Distance 3D Object Reconstruction from Static Images

chenhsuanlin/signed-distance-SRN NeurIPS 2020

Dense 3D object reconstruction from a single image has recently witnessed remarkable advances, but supervising neural networks with ground-truth 3D shapes is impractical due to the laborious process of creating paired image-shape datasets.

Share With Thy Neighbors: Single-View Reconstruction by Cross-Instance Consistency

monniert/unicorn 21 Apr 2022

Approaches for single-view reconstruction typically rely on viewpoint annotations, silhouettes, the absence of background, multiple views of the same instance, a template shape, or symmetry.