cmusatyalab / OpenTPOD

Open Toolkit for Painless Object Detection

OpenTPOD

Create deep learning based object detectors without writing a single line of code.

OpenTPOD is an all-in-one open-source tool for nonexperts to create custom deep neural network object detectors. It is designed to lower the barrier of entry and facilitates the end-to-end authoring workflow of custom object detection using state-of-art deep learning methods.

It provides the following features via an easy-to-use web interface.

• Training data management.
• Data annotation through seamless integration with OpenCV CVAT Labeling Tool.
• One-click training/fine-tuning of object detection deep neural networks, including SSD MobileNet, Faster RCNN Inception, and Faster RCNN ResNet, using Tensorflow (with and without GPU).
• One-click model export for inference with Tensorflow Serving.
• Extensible architecture for easy addition of new deep neural network architectures.

Demo Video

Documentation

• Motivation
• User Guide
• Installation and Administration Guide
• Developer Guide
• Thorough Description and Context in PhD Thesis Scaling Wearable Cognitive Assistance (Section 6.3)

Citations

Please cite the following thesis if you find OpenTPOD helps your research.

@phdthesis{wang2020scaling,
  title={Scaling Wearable Cognitive Assistance},
  author={Wang, Junjue},
  year={2020},
  school={CMU-CS-20-107, CMU School of Computer Science}
}

Acknowledgement

This research was supported by the National Science Foundation (NSF) under grant number CNS-1518865. Additional support was provided by Intel, Vodafone, Deutsche Telekom, Verizon, Crown Castle, Seagate, VMware, MobiledgeX, InterDigital, and the Conklin Kistler family fund.