Skip to content

First-Prize winner in the Core Track of the 2022 OpenCV AI Competition

License

Notifications You must be signed in to change notification settings

RMDLO/COCOpen-OpenCV

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

The COCOpen library is a scalable and open source method of generating a labelled dataset of colored images of any object category automatically. This dataset can be used for training object instance segmentation models for a wide range of applications like manufacturing, logistics, autonomous driving, and domestic services. COCOpen uses foreground images of single objects against blank backgrounds and background images similar to backgrounds found in the target deployment environment as input. The library annotates each object with its unique category identification number, bounding box, and instance segmentation mask in the Microsoft Common Objects in Context (COCO) format [1,2]. The COCOpen-OpenCV repository uses COCOpen to generate a dataset of images comprising instances of wires and networking devices, visualizes the annotations for these images, trains an instance segmentation model, and demonstrates inference with the trained model [3,4].

This is an example COCOpen-produced synthetic image containing multiple objects of interest superimposed on a randomly selected background. It visualizes ground truth instance segmentation mask, object category, and bounding box labels.

[Get the code] [Watch Our Demonstration Video]

Get Started

See installation instructions to learn about configuring a dataset and installing dependencies.

Run an Example

See example run to see a demonstration automatically generating a simple dataset of ethernet cables and ethernet devices with category, bounding box, and instance segmentation mask annotations in the COCO format.

Learn More

To learn more about COCOpen, click here.

References

[1] OpenCV, "Open Source Computer Vision Library," (2015). [website]

[2] T. Lin, M. Maire, S. Belongie, J. Hays, P. Perona, D. Ramanan, P. Dollár, and C. Lawrence Zitnick, "Microsoft COCO: Common Objects in Context," in Eur. Conf. Comput. Vis. (ECCV), Sep. 2014, pp. 740-755. [paper]

[3] K. He, G. Gkioxari, P. Dollár, and R. Girshick, "Mask R-CNN," in IEEE Int. Conf. Comput. Vis. Pattern Recognit. (CVPR), Oct. 2017, pp. 2980-2988. [paper]

[4] G. Ghiasi, Y. Cui, A. Srinivas, R. Qian, T. Lin, E. Cubuk, Q. Le, and B. Zoph, "Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation," in IEEE Int. Conf. Comput. Vis. Pattern Recognit. (CVPR), June 2020, pp. 9796-9805. [paper]