Open source from Alibaba.

README.md

@@ -0,0 +1,93 @@
+
+
+# UniFuse (RAL+ICRA2021)
+
+Office source code of paper **UniFuse: Unidirectional Fusion for 360$^\circ$ Panorama Depth Estimation**, [arXiv](https://arxiv.org/abs/2102.03550), [Demo](https://youtu.be/9vm9OMksvrc)
+
+
+
+# Preparation
+
+#### Installation
+
+Environments
+
+
+* python 3.6
+* Pytorch >= 1.0.0
+* CUDA >= 9.0
+
+
+Install requirements
+
+```bash
+pip install -r requirements.txt
+```
+
+#### Datasets 
+
+Please download the preferred datasets,  i.e., [Matterport3D](https://niessner.github.io/Matterport/), [Stanford2D3D](http://3dsemantics.stanford.edu/), [3D60](https://vcl3d.github.io/3D60/) and [PanoSUNCG](https://fuenwang.ml/project/360-depth/). For Matterport3D, please preprocess it following [M3D-README.md](UniFuse/Matterport3D/README.md).
+
+
+
+# Training 
+
+#### UniFuse on Matterport3D
+
+```
+python train.py --data_path $DATA_PATH \
+-dataset matterport3d \
+--model_name Matterport3D_UniFuse \
+--batch_size 6 \
+--num_epochs 100 \
+--height 512 \
+--width 1024 \
+--imagenet_pretrained \
+--net UniFuse 
+```
+
+#### Equirectangular baseline on Matterport3D
+
+```
+python train.py --data_path $DATA_PATH \
+-dataset matterport3d \
+--model_name Matterport3D_Equi \
+--batch_size 6 \
+--num_epochs 100 \
+--height 512 \
+--width 1024 \
+--imagenet_pretrained \
+--net Equi 
+```
+
+It is similar for other datasets. 
+
+
+# Evaluation  
+
+#### Pre-trained models
+
+The pre-trained models of UniFuse for 4 datasets are available, [Matterport3D](PretrainedModels/Matterport3D_UniFuse_cee_se_b), [Stanford2D3D](PretrainedModels/Stanford2D3D_UniFuse_cee_se_b), [3D60](PretrainedModels/3D60_UniFuse_cee_se) and [PanoSUNCG](PretrainedModels/PanoSunCG_UniFuse_cee_se).
+
+#### Test on a pre-trained model
+
+```
+python evaluate.py  --data_path $DATA_PATH --dataset matterport3d --load_weights_folder $MODEL_PATH 
+```
+
+
+
+## Citation
+
+Please cite our paper if you find our work useful in your research.
+
+```
+@article{jiang2021unifuse,
+      title={UniFuse: Unidirectional Fusion for 360$^{\circ}$ Panorama Depth Estimation}, 
+      author={Hualie Jiang and Zhe Sheng and Siyu Zhu and Zilong Dong and Rui Huang},
+	  journal={IEEE Robotics and Automation Letters},
+	  year={2021},
+	  publisher={IEEE}
+}
+```
+

UniFuse/Matterport3D/README.md

@@ -0,0 +1,24 @@
+# Matterport3D Preprocessing for 360$^\circ$ Depth Estimation
+
+
+
+# Steps
+
+* Download the Matterport3D dataset
+
+* Copy *extract.sh* to the folder of Matterport3D
+
+* Open a terminal on the folder of Matterport3D and execute 
+
+  ```bash
+  chmod u+x extract.sh && ./extract.sh
+  ```
+
+* Download [PanoBasic](https://github.com/yindaz/PanoBasic)
+
+* Copy *stitching_Matterport3D.m* to PanoBasic
+
+* Modify the directories in *stitching_Matterport3D.m*,  i.e., changing **source_dir** to the folder of Matterport3D and changing **target_dir** as the output folder of panorama images and depth maps. 
+
+* Execute  *stitching_Matterport3D.m* using Matlab.
+

UniFuse/Matterport3D/extract.sh

@@ -0,0 +1,21 @@
+#!/bin/bash
+
+clear
+
+var="Extracting"
+echo $var
+
+
+for entry in `ls -d */`; do
+	echo $entry
+	cd $entry
+    for file in `ls -d *.zip`; do
+    	echo extracting $file
+    	unzip $file
+    done
+    cd $entry
+    mv * ../
+    cd ..
+    rm -r $entry
+    cd ..
+done

UniFuse/Matterport3D/stitching_Matterport3D.m

@@ -0,0 +1,20 @@
+%% stitching Matterport3D panorama images and depth  maps
+clear; close all;
+
+source_dir = '/home/wolian/disk2/Matterport';
+target_dir = '/home/wolian/disk1/Matterport3D';
+
+listing = dir(source_dir);
+listing(ismember( {listing.name}, {'.', '..'})) = [];
+
+for i = 1:length(listing)
+
+    if ~exist(strcat(target_dir, '/', listing(i).name), 'dir')
+        mkdir(strcat(strcat(target_dir, '/', listing(i).name)))
+    else
+        continue
+    end
+
+    stitch(strcat(source_dir, '/', listing(i).name), strcat(target_dir, '/', listing(i).name), listing(i).name);
+
+end

UniFuse/README.md

@@ -0,0 +1,93 @@
+
+
+# UniFuse (RAL+ICRA2021)
+
+Office source code of paper **UniFuse: Unidirectional Fusion for 360$^\circ$ Panorama Depth Estimation**, [arXiv](https://arxiv.org/abs/2102.03550), [Demo](https://youtu.be/9vm9OMksvrc)
+
+
+
+# Preparation
+
+#### Installation
+
+Environments
+
+
+* python 3.6
+* Pytorch >= 1.0.0
+* CUDA >= 9.0
+
+
+Install requirements
+
+```bash
+pip install -r requirements.txt
+```
+
+#### Datasets 
+
+Please download the preferred datasets,  i.e., [Matterport3D](https://niessner.github.io/Matterport/), [Stanford2D3D](http://3dsemantics.stanford.edu/), [3D60](https://vcl3d.github.io/3D60/) and [PanoSUNCG](https://fuenwang.ml/project/360-depth/). For Matterport3D, please preprocess it following [M3D-README.md](UniFuse/Matterport3D/README.md).
+
+
+
+# Training 
+
+#### UniFuse on Matterport3D
+
+```
+python train.py --data_path $DATA_PATH \
+-dataset matterport3d \
+--model_name Matterport3D_UniFuse \
+--batch_size 6 \
+--num_epochs 100 \
+--height 512 \
+--width 1024 \
+--imagenet_pretrained \
+--net UniFuse 
+```
+
+#### Equirectangular baseline on Matterport3D
+
+```
+python train.py --data_path $DATA_PATH \
+-dataset matterport3d \
+--model_name Matterport3D_Equi \
+--batch_size 6 \
+--num_epochs 100 \
+--height 512 \
+--width 1024 \
+--imagenet_pretrained \
+--net Equi 
+```
+
+It is similar for other datasets. 
+
+
+# Evaluation  
+
+#### Pre-trained models
+
+The pre-trained models of UniFuse for 4 datasets are available, [Matterport3D](PretrainedModels/Matterport3D_UniFuse_cee_se_b), [Stanford2D3D](PretrainedModels/Stanford2D3D_UniFuse_cee_se_b), [3D60](PretrainedModels/3D60_UniFuse_cee_se) and [PanoSUNCG](PretrainedModels/PanoSunCG_UniFuse_cee_se).
+
+#### Test on a pre-trained model
+
+```
+python evaluate.py  --data_path $DATA_PATH --dataset matterport3d --load_weights_folder $MODEL_PATH 
+```
+
+
+
+## Citation
+
+Please cite our paper if you find our work useful in your research.
+
+```
+@article{jiang2021unifuse,
+      title={UniFuse: Unidirectional Fusion for 360$^{\circ}$ Panorama Depth Estimation}, 
+      author={Hualie Jiang and Zhe Sheng and Siyu Zhu and Zilong Dong and Rui Huang},
+	  journal={IEEE Robotics and Automation Letters},
+	  year={2021},
+	  publisher={IEEE}
+}
+```
+