IRS

IRS: A Large Synthetic Indoor Robotics Stereo Dataset for Disparity and Surface Normal Estimation

Introduction

IRS is an open dataset for indoor robotics vision tasks, especially disparity and surface normal estimation. It contains totally 103,316 samples covering a wide range of indoor scenes, such as home, office, store and restaurant.


Left image	Right image

Disparity map	Surface normal map

Overview of IRS

Rendering Characteristic	Options
indoor scene class	home(31145), office(43417), restaurant(22058), store(6696)
object class	desk, chair, sofa, glass, mirror, bed, bedside table, lamp, wardrobe, etc.
brightness	over-exposure(>1300), darkness(>1700)
light behavior	bloom(>1700), lens flare(>1700), glass transmission(>3600), mirror reflection(>3600)

We give some sample of different indoor scene characteristics as follows.


Home	Office	Restaurant

Normal light	Over exposure	Darkness

Glass	Mirror	Metal

Network Structure of DispNormNet

We design a novel network, namely DispNormNet, to estimate the disparity map and surface normal map together of the input stereo images. DispNormNet is comprised of two modules, DispNetC and NormNetDF. DispNetC is identical to that in this paper and produces the disparity map. NormNetDF produces the normal map and is similar to DispNetS. "DF" indicates disparity feature fusion, which we found important to produce accurate surface normal maps.

DispNormNet

Paper

Q. Wang^*,1, S. Zheng^*,1, Q. Yan^*,2, F. Deng², K. Zhao^†,1, X. Chu^†,1.

IRS : A Large Synthetic Indoor Robotics Stereo Dataset for Disparity and Surface Normal Estimation. [preprint]

* indicates equal contribution. † indicates corresponding authors.
¹Department of Computer Science, Hong Kong Baptist University. ²School of Geodesy and Geomatics, Wuhan University.

Download

You can use the following BaiduYun link to download our dataset. More download links, including Google Drive and OneDrive, will be provided soon.

BaiduYun: https://pan.baidu.com/s/1VKVVdljNdhoyJ8JdQUCwKQ

Google Drive: https://drive.google.com/drive/folders/1GoMbbiAJuIE1ArhdR4Uj__HzScTvj6NP?usp=sharing

Video Demonstration

Usage

Dependencies

Python2.7
PyTorch(1.2.0)
torchvision 0.2.0 (higher version may cause issues)
Cuda 10 (https://developer.nvidia.com/cuda-downloads)

Install

Use the following commands to install the environment in Linux

cd layers_package
./install.sh

# install OpenEXR (https://www.openexr.com/)
sudo apt-get update
sudo apt-get install openexr

Dataset

Download IRS dataset from https://pan.baidu.com/s/1VKVVdljNdhoyJ8JdQUCwKQ (BaiduYun).
Extract zip files and put them in correct folder:

---- pytorch-dispnet ---- data ---- IRSDataset ---- Home
                                                |-- Office
                                                |-- Restaurant
                                                |-- Store

Train

There are configurations for train in "exp_configs" folder. You can create your own configuration file as samples.
As an example, following configuration can be used to train a DispNormNet on IRS dataset:

/exp_configs/dispnormnet.conf

net=dispnormnet
loss=loss_configs/dispnetcres_irs.json
outf_model=models/${net}-irs
logf=logs/${net}-irs.log

lr=1e-4
devices=0,1,2,3

dataset=irs #sceneflow, irs, sintel
trainlist=lists/IRSDataset_TRAIN.list
vallist=lists/IRSDataset_TEST.list

startR=0
startE=0
endE=10
batchSize=16
maxdisp=-1
model=none

Then, the configuration should be specified in the "train.sh"

/train.sh

dnn="${dnn:-dispnormnet}"
source exp_configs/$dnn.conf

python main.py --cuda --net $net --loss $loss --lr $lr \
               --outf $outf_model --logFile $logf \
               --devices $devices --batch_size $batchSize \
               --dataset $dataset --trainlist $trainlist --vallist $vallist \
               --startRound $startR --startEpoch $startE --endEpoch $endE \
               --model $model \
               --maxdisp $maxdisp \
               --manualSeed 1024 \

Lastly, use the following command to start a train

./train.sh

Evaluation

There is a script for evaluation with a model from a train

/detech.sh

dataset=irs
net=dispnormnet

model=models/dispnormnet-irs/model_best.pth
outf=detect_results/${net}-${dataset}/

filelist=lists/IRSDataset_TEST.list
filepath=data

CUDA_VISIBLE_DEVICES=0 python detecter.py --model $model --rp $outf --filelist $filelist --filepath $filepath --devices 0 --net ${net} --disp-on --norm-on

Use the script in your configuration, and then get result in detect_result folder.

Disparity results are saved in png format as default.
Normal results are saved in exr format as default.

If you want to change the output format, you need to modify "detecter.py" and use save function as follow

# png
skimage.io.imsave(filepath, image)

# pfm
save_pfm(filepath, data)

# exr
save_exr(data, filepath)

EXR Viewer

For viewing files in exr format, we recommand a free software

RenderDoc

Contact

Please contact us at qiangwang@comp.hkbu.edu.hk if you have any question.

ling-zzZ / IRS

IRS