GuoleiSun / Countseg
Official code for "Object counting and instance segmentation with image-level supervision", in CVPR 2019 and TPAMI 2020
Stars: β 129
Programming Languages
python
139335 projects - #7 most used programming language
Projects that are alternatives of or similar to Countseg
Sota Point Cloud
π₯Deep Learning for 3D Point Clouds (IEEE TPAMI, 2020)
Stars: β 778 (+503.1%)
Mutual labels: instance-segmentation
Panet
PANet for Instance Segmentation and Object Detection
Stars: β 1,170 (+806.98%)
Mutual labels: instance-segmentation
Bmaskr Cnn
Boundary-preserving Mask R-CNN (ECCV 2020)
Stars: β 116 (-10.08%)
Mutual labels: instance-segmentation
Acis
Actor-Critic Instance Segmentation (CVPR 2019)
Stars: β 15 (-88.37%)
Mutual labels: instance-segmentation
Data Science Bowl 2018
End-to-end one-class instance segmentation based on U-Net architecture for Data Science Bowl 2018 in Kaggle
Stars: β 56 (-56.59%)
Mutual labels: instance-segmentation
Awesome Referring Image Segmentation
π A collection of papers about Referring Image Segmentation.
Stars: β 91 (-29.46%)
Mutual labels: instance-segmentation
Polyrnn Pp
Inference Code for Polygon-RNN++ (CVPR 2018)
Stars: β 704 (+445.74%)
Mutual labels: instance-segmentation
Deepco3
[CVPR19] DeepCO3: Deep Instance Co-segmentation by Co-peak Search and Co-saliency (Oral paper)
Stars: β 127 (-1.55%)
Mutual labels: instance-segmentation
Maskrcnn Modanet
A Mask R-CNN Keras implementation with Modanet annotations on the Paperdoll dataset
Stars: β 59 (-54.26%)
Mutual labels: instance-segmentation
Lanenet Lane Detection
Unofficial implemention of lanenet model for real time lane detection using deep neural network model https://maybeshewill-cv.github.io/lanenet-lane-detection/
Stars: β 1,690 (+1210.08%)
Mutual labels: instance-segmentation
Gcnet
GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond
Stars: β 945 (+632.56%)
Mutual labels: instance-segmentation
Yolact fcos
YOLACT: Real-time Instance Segmentation on the FCOS detector (without bbox cropping), achives 35.2mAP on coco val
Stars: β 56 (-56.59%)
Mutual labels: instance-segmentation
Condinst
Conditional Convolutions for Instance Segmentation, achives 37.1mAP on coco val
Stars: β 101 (-21.71%)
Mutual labels: instance-segmentation
Cell Detr
Official and maintained implementation of the paper Attention-Based Transformers for Instance Segmentation of Cells in Microstructures [BIBM 2020].
Stars: β 26 (-79.84%)
Mutual labels: instance-segmentation
Ac Fpn
Implement of paper γAttention-guided Context Feature Pyramid Network for Object Detectionγ
Stars: β 117 (-9.3%)
Mutual labels: instance-segmentation
Labelme
Image Polygonal Annotation with Python (polygon, rectangle, circle, line, point and image-level flag annotation).
Stars: β 7,742 (+5901.55%)
Mutual labels: instance-segmentation
Seg By Interaction
Unsupervised instance segmentation via active robot interaction
Stars: β 78 (-39.53%)
Mutual labels: instance-segmentation
Swa object detection
SWA Object Detection
Stars: β 128 (-0.78%)
Mutual labels: instance-segmentation
Mask rcnn pytorch
Mask R-CNN for object detection and instance segmentation on Pytorch
Stars: β 123 (-4.65%)
Mutual labels: instance-segmentation
Official code for Object Counting and Instance Segmentation with Image-level Supervision, in CVPR 2019 and TPAMI 2020
Paper (conference version) Paper (journal version) Supp Presentation Poster
This code is built by Guolei Sun. For questions, please contact [email protected]
Requirements:
- System: ubuntu 16.04.
- NVIDIA GPU + CUDA CuDNN
- Python>=3.6
- Pytorch version 0.4
- Jupyter Notebook and ipywidgets
- Other common packages: numpy, scipy, and so on. Please refer to environment.yml.
Installation:
This respority uses some functions from PRM.
- Download this respority and unzip it. Make sure that the folders look like this:
CountSeg
βββ Nest-pytorch
βββ ...
βββ PRM-pytorch
βββ ...
βββ ...
- Go inside to "CountSeg" folder by "cd path/CountSeg", where path is where you store CountSeg in your computer.
- Install Nest, a flexible tool for building and sharing deep learning modules, created by Yanzhao
pip install git+https://github.com/ZhouYanzhao/Nest.git
- Install PRM via Nest's CLI tool
Validate the installation by "nest module list --filter prm", you should see something like this.nest module install ./PRM-pytorch prm
If you get some error, it is because that you miss some packages. Install them and do the validation again until you can get something like above## Output: # # 3 Nest modules found. #[0] prm.fc_resnet50 (1.0.0) #[1] prm.peak_response_mapping (1.0.0) #[2] prm.prm_visualize (1.0.0)
- Install Nest's build-in Pytorch modules
nest module install ./Nest-pytorch pytorch
Dataset
- For Pascal, download dataset by simply running script in CountSeg/dataset folder.
Before running the script, do not forget to change the save_path_2007 and save_path_2012 in pascal_download.sh to the path where you want to save pascal 2007 and 2012 dataset.cd path/CountSeg/dataset chmod +x pascal_download.sh ./pascal_download.sh
- For COCO dataset, please download dataset from COCO
Demo
Go to "CountSeg/demo" and run demo.
Test
To reproduce the results reproted in the paper.
- Pascal 2007 counting
Then open eval_counting_pascal07.ipynb and simply run each cell inside it. Make sure you change the data path in eval_counting_pascal07.ipynb.cd path/CountSeg jupyter notebook
- COCO 2014 counting. Open jupyter notebook and run eval_counting_coco14.ipynb. Make sure you change the data path in eval_counting_coco14.ipynb.
Training
- Pascal 2007 and COCO 2014.
Then open main-pascal.ipynb or main-coco.ipynb to do training on Pascal or COCO, respectively. Make sure you change the data path in config_counting_pascal07.yml and config_counting_coco14.yml.cd path/CountSeg/experiments jupyter notebook
Citation
If you find the code useful for your research, please cite:
@INPROCEEDINGS{cholakkal_sun2019object,
author = {Cholakkal, Hisham and Sun, Guolei and Khan, Fahad Shahbaz and Shao, Ling},
title = {Object Counting and Instance Segmentation with Image-level Supervision},
booktitle = {CVPR},
year = {2019}
}
@article{cholakkal_sun2020towards,
title={Towards Partial Supervision for Generic Object Counting in Natural Scenes},
author={Cholakkal, Hisham and Sun, Guolei and Khan, Salman and Khan, Fahad Shahbaz and Shao, Ling and Gool, Luc Van},
journal={IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)},
year={2020}
}
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].