If you use this code, please cite the following paper:

Lin S, Cheng R, Wang K, Yang K. Visual Localizer: Outdoor Localization Based on ConvNet Descriptor and Global Optimization for Visually Impaired Pedestrians. Sensors. 2018; 18(8):2476. https://doi.org/10.3390/s18082476

The code utilizes the ideas of Visual Localizer, which uses the compressed feature maps of ConvNet as image descriptors and uses Network Flow as the global optiomization tool.

Denpendency:

• or-tools
• OpenCV 4