GitPlanet

Cheap and reliable Node.js hosting starts at $3/month, and $1/month static HTML hosting

Created with love in Canada, visit hostnodejs.com today

Feel like to post an Ad? Learn Details
All Projects → rehg-lab → 3DShapeGen

rehg-lab / 3DShapeGen

Licence: MIT license
Code for 3D Reconstruction of Novel Object Shapes from Single Images paper

Programming Languages

python
139335 projects - #7 most used programming language
C++
36643 projects - #6 most used programming language
c
50402 projects - #5 most used programming language
cython
566 projects
Mako
254 projects

Labels

3d-reconstructiongeneralizationshape-from-image

Projects that are alternatives of or similar to 3DShapeGen

New-View-Synthesis
Collecting papers about new view synthesis
Stars: ✭ 437 (+375%)
Mutual labels:  3d-reconstruction
ShapeFormer
Official repository for the ShapeFormer Project
Stars: ✭ 97 (+5.43%)
Mutual labels:  3d-reconstruction
SpinNet
[CVPR 2021] SpinNet: Learning a General Surface Descriptor for 3D Point Cloud Registration
Stars: ✭ 181 (+96.74%)
Mutual labels:  generalization
retrieval-fuse
[ICCV21] Code for "RetrievalFuse: Neural 3D Scene Reconstruction with a Database"
Stars: ✭ 69 (-25%)
Mutual labels:  3d-reconstruction
StructuredLight
Creating a 3D reconstruction of an object using multiple images
Stars: ✭ 42 (-54.35%)
Mutual labels:  3d-reconstruction
MVSNet pl
MVSNet: Depth Inference for Unstructured Multi-view Stereo using pytorch-lightning
Stars: ✭ 49 (-46.74%)
Mutual labels:  3d-reconstruction
semantic-tsdf
Semantic-TSDF for Self-driving Static Scene Reconstruction
Stars: ✭ 14 (-84.78%)
Mutual labels:  3d-reconstruction
ICON
ICON: Implicit Clothed humans Obtained from Normals (CVPR 2022)
Stars: ✭ 641 (+596.74%)
Mutual labels:  3d-reconstruction
pyRANSAC-3D
A python tool for fitting primitives 3D shapes in point clouds using RANSAC algorithm
Stars: ✭ 253 (+175%)
Mutual labels:  3d-reconstruction
modules
The official repository for our paper "Are Neural Nets Modular? Inspecting Functional Modularity Through Differentiable Weight Masks". We develop a method for analyzing emerging functional modularity in neural networks based on differentiable weight masks and use it to point out important issues in current-day neural networks.
Stars: ✭ 25 (-72.83%)
Mutual labels:  generalization
over-parametrization
Computing various norms/measures on over-parametrized neural networks
Stars: ✭ 46 (-50%)
Mutual labels:  generalization
softpool
SoftPoolNet: Shape Descriptor for Point Cloud Completion and Classification - ECCV 2020 oral
Stars: ✭ 62 (-32.61%)
Mutual labels:  3d-reconstruction
Generalizing-Lottery-Tickets
This repository contains code to replicate the experiments given in NeurIPS 2019 paper "One ticket to win them all: generalizing lottery ticket initializations across datasets and optimizers"
Stars: ✭ 48 (-47.83%)
Mutual labels:  generalization
PaiConvMesh
Official repository for the paper "Learning Local Neighboring Structure for Robust 3D Shape Representation"
Stars: ✭ 19 (-79.35%)
Mutual labels:  3d-reconstruction
LPDC-Net
CVPR2021 paper "Learning Parallel Dense Correspondence from Spatio-Temporal Descriptorsfor Efficient and Robust 4D Reconstruction"
Stars: ✭ 27 (-70.65%)
Mutual labels:  3d-reconstruction
paper-notes
ML/DL/RL paper notes
Stars: ✭ 21 (-77.17%)
Mutual labels:  generalization
RfDNet
Implementation of CVPR'21: RfD-Net: Point Scene Understanding by Semantic Instance Reconstruction
Stars: ✭ 150 (+63.04%)
Mutual labels:  3d-reconstruction
Face-X
Demonstration of different algorithms and operations on faces. Join the Discord channel for discussion.
Stars: ✭ 226 (+145.65%)
Mutual labels:  3d-reconstruction
dynamic plane convolutional onet
[WACV 2021] Dynamic Plane Convolutional Occupancy Networks
Stars: ✭ 25 (-72.83%)
Mutual labels:  3d-reconstruction
extrapolate
generalize counter-examples of property-based testing
Stars: ✭ 13 (-85.87%)
Mutual labels:  generalization
View All Similar Projects ➔

3D Reconstruction of Novel Object Shapes from Single Images

preview

In this work we present a comprehensive exploration of generalization to unseen shapes in single-view 3D reconstruction. We introduce SDFNet, an architecture combining 2.5D sketch estimation with a continuous shape regressor for signed distance functions of objects. We show new findings on the impact of rendering variability and adopting a 3-DOF VC (3 Degree-of-Freedom Viewer Centered) coordinate representation on generalization to object shapes not seen during training. Our model can generalize to objects of unseen categories, and to objects from a significantly different shape dataset. Link to our paper and link to our project webpage.

This repository consists of the code for rendering, training and evaluating SDFNet as well as baseline method Occupancy Networks. Code to repoduce results for baseline method GenRe can be found here.

Training and evaluating SDFNet and OccNet

Follow instructions in SDFNet README

Training and evaluating GenRe

Follow instruction in GenRe README

Rendering

Follow instructions in Rendering README

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].