Code release for BlockGAN: Learning 3D Object-aware Scene Representations from Unlabelled Images

Related tags

Deep LearningBlockGAN
Overview

BlockGAN

Code release for BlockGAN: Learning 3D Object-aware Scene Representations from Unlabelled Images

trans add

BlockGAN: Learning 3D Object-aware Scene Representations from Unlabelled Images
Thu Nguyen-Phuoc, Chrisian Richardt, Long Mai, Yong-liang Yang, Niloy Mitra

Dataset

Please contact Thu Nguyen-Phuoc for datasets.

Training

  • To run the training of BlockGAN
python main.py ./config_synthetic.json --dataset Chair --input_fname_pattern ".png" 

python main.py ./config_real.json --dataset Car --input_fname_pattern ".jpg"

Help with config.json

image_path:
			Full path to the dataset directory.
gpu:
			Index number of the GPU to use. Default: 0.
batch_size:
			Batch size. Defaults is 32.
max_epochs:
			Number of epochs to train. Defaults is 50.
epoch_step:
			Number of epochs to train before starting to decrease the learning rate. Default is 25.
z_dim:
			Dimension of the noise vector. Defaults is 90.
z_dim2:
			Dimension of the noise vector. Defaults is 30.			
d_eta:
			Learning rate of the discriminator.Default is 0.0001
g_eta:
			Learning rate of the generator.Default is 0.0001
reduce_eta:
			Reduce learning rate during training.Default is False
D_update:
			Number of updates for the Discriminator for every training step.Default is 1.
G_update:
			Number of updates for the Generator for every training step.Default is 2.
beta1:
			Beta 1 for the Adam optimiser. Default is 0.5
beta2:
			Beta 2 for the Adam optimiser. Default is 0.999
discriminator:
			Name of the discriminator to use. 
generator:
			Name of the generator to use. 
view_func:
			Name of the view sampling function to use.
skew_func:
			Name of the perspective skew function to use.
train_func:
			Name of the train function to use.
build_func:
			Name of the build function to use.
style_disc:
			Use Style discriminator. Useful for training images at 128.
sample_z:
			Distribution to sample the noise fector. Default is "uniform".
add_D_noise:
			Add noise to the input of the discriminator. Default is "false".
DStyle_lambda:
			Lambda for the style discriminator loss. Default is 1.0
ele_low:
    		        Default is 70.
ele_high:
			Default is 110.
azi_low:
			Default is 0.
azi_high:
			Default is 360.
scale_low:
			Default is 1.0
scale_high:
			Default is 1.0
x_low:
			Default is 0.
x_high:
			Default is 0.
y_low:
			Default is 0.
y_high:
			Default is 0.
z_low:
			Default is 0.
z_high:
			Default is 0.
with_translation:
			To use translation in 3D transformation. Default is "true".
with_scale:
			To use scaling in 3D transformation. Default is "true".
focal_length:
			Camera parameter. Default is 35.
sensor_size:
			Camera parameter. Default is 32.
camera_dist:
			Camera distance. Default is 11.
new_size:
			Voxel grid size. Default is 16.	
size:
			Voxel grid size. Default is 16.	
output_dir: 
			Full path to the output directory.

Citation

If you use this code for your research, please cite our paper

@inproceedings{BlockGAN2020,
  title={ BlockGAN: Learning 3D Object-aware Scene Representations from Unlabelled Images  },
  author={Nguyen-Phuoc, Thu and Richardt, Christian and Mai, Long and Yang, Yong-Liang and Mitra, Niloy},
  booktitle =  {Advances in Neural Information Processing Systems 33},
 month = {Nov},
 year = {2020}
}
A python tutorial on bayesian modeling techniques (PyMC3)

Bayesian Modelling in Python Welcome to "Bayesian Modelling in Python" - a tutorial for those interested in learning how to apply bayesian modelling t

Mark Regan 2.4k Jan 06, 2023
This is an example of object detection on Micro bacterium tuberculosis using Mask-RCNN

Mask-RCNN on Mycobacterium tuberculosis This is an example of object detection on Mycobacterium Tuberculosis using Mask RCNN. Implement of Mask R-CNN

Jun-En Ding 1 Sep 16, 2021
Pytorch implementation of the paper SPICE: Semantic Pseudo-labeling for Image Clustering

SPICE: Semantic Pseudo-labeling for Image Clustering By Chuang Niu and Ge Wang This is a Pytorch implementation of the paper. (In updating) SOTA on 5

Chuang Niu 154 Dec 15, 2022
A PyTorch Implementation of "Watch Your Step: Learning Node Embeddings via Graph Attention" (NeurIPS 2018).

Attention Walk ⠀⠀ A PyTorch Implementation of Watch Your Step: Learning Node Embeddings via Graph Attention (NIPS 2018). Abstract Graph embedding meth

Benedek Rozemberczki 303 Dec 09, 2022
RL-driven agent playing tic-tac-toe on starknet against challengers.

tictactoe-on-starknet RL-driven agent playing tic-tac-toe on starknet against challengers. GUI reference: https://pythonguides.com/create-a-game-using

21 Jul 30, 2022
Breaching - Breaching privacy in federated learning scenarios for vision and text

Breaching - A Framework for Attacks against Privacy in Federated Learning This P

Jonas Geiping 139 Jan 03, 2023
Discord bot for notifying on github events

Git-Observer Discord bot for notifying on github events ⚠️ This bot is meant to write messages to only one channel (implementing this for multiple pro

ilu_vatar_ 0 Apr 19, 2022
EPSANet:An Efficient Pyramid Split Attention Block on Convolutional Neural Network

EPSANet:An Efficient Pyramid Split Attention Block on Convolutional Neural Network This repo contains the official Pytorch implementaion code and conf

Hu Zhang 175 Jan 07, 2023
202 Jan 06, 2023
[CVPR 2021] 'Searching by Generating: Flexible and Efficient One-Shot NAS with Architecture Generator'

[CVPR2021] Searching by Generating: Flexible and Efficient One-Shot NAS with Architecture Generator Overview This is the entire codebase for the paper

35 Dec 01, 2022
Official project repository for 'Normality-Calibrated Autoencoder for Unsupervised Anomaly Detection on Data Contamination'

NCAE_UAD Official project repository of 'Normality-Calibrated Autoencoder for Unsupervised Anomaly Detection on Data Contamination' Abstract In this p

Jongmin Andrew Yu 2 Feb 10, 2022
Deep GPs built on top of TensorFlow/Keras and GPflow

GPflux Documentation | Tutorials | API reference | Slack What does GPflux do? GPflux is a toolbox dedicated to Deep Gaussian processes (DGP), the hier

Secondmind Labs 107 Nov 02, 2022
Adversarial Self-Defense for Cycle-Consistent GANs

Adversarial Self-Defense for Cycle-Consistent GANs This is the official implementation of the CycleGAN robust to self-adversarial attacks used in pape

Dina Bashkirova 10 Oct 10, 2022
Code for T-Few from "Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning"

T-Few This repository contains the official code for the paper: "Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learni

220 Dec 31, 2022
Get a Grip! - A robotic system for remote clinical environments.

Get a Grip! Within clinical environments, sterilization is an essential procedure for disinfecting surgical and medical instruments. For our engineeri

Jay Sharma 1 Jan 05, 2022
CTRL-C: Camera calibration TRansformer with Line-Classification

CTRL-C: Camera calibration TRansformer with Line-Classification This repository contains the official code and pretrained models for CTRL-C (Camera ca

57 Nov 14, 2022
In the AI for TSP competition we try to solve optimization problems using machine learning.

AI for TSP Competition Goal In the AI for TSP competition we try to solve optimization problems using machine learning. The competition will be hosted

Paulo da Costa 11 Nov 27, 2022
Code for the preprint "Well-classified Examples are Underestimated in Classification with Deep Neural Networks"

This is a repository for the paper of "Well-classified Examples are Underestimated in Classification with Deep Neural Networks" The implementation and

LancoPKU 25 Dec 11, 2022
Numenta published papers code and data

Numenta research papers code and data This repository contains reproducible code for selected Numenta papers. It is currently under construction and w

Numenta 293 Jan 06, 2023
Official Pytorch implementation of "Learning to Estimate Robust 3D Human Mesh from In-the-Wild Crowded Scenes", CVPR 2022

Learning to Estimate Robust 3D Human Mesh from In-the-Wild Crowded Scenes / 3DCrowdNet News 💪 3DCrowdNet achieves the state-of-the-art accuracy on 3D

Hongsuk Choi 113 Dec 21, 2022