Unseen Object Amodal Instance Segmentation (UOAIS)

Seunghyeok Back, Joosoon Lee, Taewon Kim, Sangjun Noh, Raeyoung Kang, Seongho Bak, Kyoobin Lee

This repository contains source codes for the paper "Unseen Object Amodal Instance Segmentation via Hierarchical Occlusion Modeling." (ICRA 2022)

[Paper] [ArXiv] [Project Website] [Video]

Updates & TODO Lists

(2021.09.26) UOAIS-Net has been released
(2021.11.15) Inference codes for kinect azure and OSD dataset.
(2021.11.22) ROS nodes for kinect azure and realsense D435
(2021.12.22) Train and evaluation codes on OSD and OCID dataset + OSD-Amodal annotation

Getting Started

Environment Setup

Tested on Titan RTX with python 3.7, pytorch 1.8.0, torchvision 0.9.0, CUDA 10.2 / 11.1 and detectron2 v0.5 / v0.6

Download source codes and checkpoints

git clone https://github.com/gist-ailab/uoais.git
cd uoais
mkdir output

Download checkpoints at GDrive
Move the R50_depth_mlc_occatmask_hom_concat and R50_rgbdconcat_mlc_occatmask_hom_concat to the output folder.
Move the rgbd_fg.pth to the foreground_segmentation folder
Set up a python environment

conda create -n uoais python=3.8
conda activate uoais
pip install torch torchvision 
pip install shapely torchfile opencv-python pyfastnoisesimd rapidfuzz termcolor

Install detectron2
Build custom AdelaiDet inside this repo (at the uoais folder)

python setup.py build develop

Run on sample OSD dataset

# UOAIS-Net (RGB-D) + CG-Net (foreground segmentation)
python tools/run_on_OSD.py --use-cgnet --dataset-path ./sample_data --config-file configs/R50_rgbdconcat_mlc_occatmask_hom_concat.yaml
# UOAIS-Net (depth) + CG-Net (foreground segmentation)
python tools/run_on_OSD.py --use-cgnet --dataset-path ./sample_data  --config-file configs/R50_depth_mlc_occatmask_hom_concat.yaml
# UOAIS-Net (RGB-D)
python tools/run_on_OSD.py --dataset-path ./sample_data --config-file configs/R50_rgbdconcat_mlc_occatmask_hom_concat.yaml
# UOAIS-Net (depth)
python tools/run_on_OSD.py --dataset-path ./sample_data --config-file configs/R50_depth_mlc_occatmask_hom_concat.yaml

Run with ROS

Realsense D-435 (realsense-ros is required.)

# launch realsense2 driver
roslaunch realsense2_camera rs_aligned_depth.launch
# launch uoais node
roslaunch uoais uoais_rs_d435.launch 
# or you can use rosrun
rosrun uoais uoais_node.py _mode:="topic"

Azure Kinect (Azure_kinect_ROS_Driver is required)

# launch azure kinect driver
roslaunch azure_kinect_ros_driver driver.launch
# launch uoais node
roslaunch uoais uoais_k4a.launch

Topics & service

/uoais/vis_img (sensor_msgs/Image): visualization results
/uoais/results (uoais/UOAISResults): UOAIS-Net predictions (mode:=topic)
/get_uoais_results (uoais/UOAISRequest): UOAIS-Net predictions (mode:=service)

Parameters

mode (string): running mode of ros node (topic or service)
rgb (string): topic name of the input rgb
depth (string): topic name of the input depth
camera_info (string): topic name of the input camera info
use_cgnet (bool): use CG-Net [1] for foreground segmentation or not
use_planeseg (bool): use RANSAC for plane segmentation or not
ransac_threshold (float): max distance a point can be from the plane model

Run without ROS

Realsense D-435 (librealsense and pyrealsense2 are required.)

# UOAIS-Net (RGB-D) + CG-Net (foreground segmentation)
python tools/rs_demo.py --use-cgnet --config-file configs/R50_rgbdconcat_mlc_occatmask_hom_concat.yaml
UOAIS-Net (depth) + CG-Net (foreground segmentation)
python tools/rs_demo.py --use-cgnet --config-file configs/R50_depth_mlc_occatmask_hom_concat.yaml
# UOAIS-Net (RGB-D)
python tools/rs_demo.py --config-file configs/R50_rgbdconcat_mlc_occatmask_hom_concat.yaml
# UOAIS-Net (depth)
python tools/rs_demo.py --config-file configs/R50_depth_mlc_occatmask_hom_concat.yaml

Azure Kinect (Azure-Kinect-Sensor-SDK and pyk4a are required.)

# UOAIS-Net (RGB-D) + CG-Net (foreground segmentation)
python tools/k4a_demo.py --use-cgnet --config-file configs/R50_rgbdconcat_mlc_occatmask_hom_concat.yaml
UOAIS-Net (depth) + CG-Net (foreground segmentation)
python tools/k4a_demo.py --use-cgnet --config-file configs/R50_depth_mlc_occatmask_hom_concat.yaml
# UOAIS-Net (RGB-D)
python tools/k4a_demo.py --config-file configs/R50_rgbdconcat_mlc_occatmask_hom_concat.yaml
# UOAIS-Net (depth)
python tools/k4a_demo.py --config-file configs/R50_depth_mlc_occatmask_hom_concat.yaml

Train & Evaluation

Dataset Preparation

Download UOAIS-Sim.zip and OSD-Amodal-annotations.zip at GDrive
Download OSD-0.2-depth.zip at OSD. [2]
Download OCID dataset at OCID. [3]
Extract the downloaded datasets and organize the folders as follows

uoais
├── output
└── datasets
       ├── OCID-dataset # for evaluation on indoor scenes
       │     └──ARID10
       │     └──ARID20
       │     └──YCB10
       ├── OSD-0.20-depth # for evaluation on tabletop scenes
       │     └──amodal_annotation # OSD-amodal
       │     └──annotation
       │     └──disparity
       │     └──image_color
       │     └──occlusion_annotation # OSD-amodal
       └── UOAIS-Sim # for training
              └──annotations
              └──train
              └──val

Train on UOAIS-Sim

# UOAIS-Net (RGB-D) 
python train_net.py --config-file configs/R50_rgbdconcat_mlc_occatmask_hom_concat.yaml
# UOAIS-Net (depth) 
python train_net.py --config-file configs/R50_depth_mlc_occatmask_hom_concat.yaml

Evaluation on OSD dataset

# UOAIS-Net (RGB-D) + CG-Net (foreground segmentation)
python eval/eval_on_OSD.py --config-file configs/R50_rgbdconcat_mlc_occatmask_hom_concat.yaml --use-cgnet
# UOAIS-Net (depth) + CG-Net (foreground segmentation)
python eval/eval_on_OSD.py --config-file configs/R50_depth_mlc_occatmask_hom_concat.yaml --use-cgnet

This code evaluates the UOAIS-Net that was trained on a single seed (7), thus the metrics from this code and the paper (an average of seeds 7, 77, 777) can be different.

Evaluation on OCID dataset

# UOAIS-Net (RGB-D)
python eval/eval_on_OCID.py --config-file configs/R50_rgbdconcat_mlc_occatmask_hom_concat.yaml
# UOAIS-Net (depth)
python eval/eval_on_OCID.py --config-file configs/R50_depth_mlc_occatmask_hom_concat.yaml

Visualization on OSD dataset

python tools/run_on_OSD.py --use-cgnet --config-file configs/R50_rgbdconcat_mlc_occatmask_hom_concat.yaml

License

The source code of this repository is released only for academic use. See the license file for details.

Notes

The codes of this repository are built upon the following open sources. Thanks to the authors for sharing the code!

Instance segmentation based on Detectron2 and AdelaiDet
Evaluation codes are modified from [4] UCN and [5] VRSP-Net.
Synthetic dataset is generated with [6] BlenderProc

Citation

If you use our work in a research project, please cite our work:

@inproceedings{back2022unseen,
  title={Unseen object amodal instance segmentation via hierarchical occlusion modeling},
  author={Back, Seunghyeok and Lee, Joosoon and Kim, Taewon and Noh, Sangjun and Kang, Raeyoung and Bak, Seongho and Lee, Kyoobin},
  booktitle={2022 International Conference on Robotics and Automation (ICRA)},
  pages={5085--5092},
  year={2022},
  organization={IEEE}
}

References

[1] SUN, Yao, et al. Cg-net: Conditional gis-aware network for individual building segmentation in vhr sar images. IEEE Transactions on Geoscience and Remote Sensing, 2021, 60: 1-15.
[2] Richtsfeld, Andreas, et al. "Segmentation of unknown objects in indoor environments." 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems. IEEE, 2012.
[3] Suchi, Markus, et al. "EasyLabel: a semi-automatic pixel-wise object annotation tool for creating robotic RGB-D datasets." 2019 International Conference on Robotics and Automation (ICRA). IEEE, 2019.
[4] Xiang, Yu, et al. "Learning rgb-d feature embeddings for unseen object instance segmentation." Conference on Robot Learning (CoRL). 2020.
[5] Xiao, Yuting, et al. "Amodal Segmentation Based on Visible Region Segmentation and Shape Prior." Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 35. No. 4. 2021.
[6] DENNINGER, Maximilian, et al. Blenderproc. arXiv preprint arXiv:1911.01911, 2019.