This is the official code for the TWMM. TWMM (Template matching with Weights, Multilevel local max-pooling, and Max index backtracking) is a registration method for UAV thermal infrared and visible images taken by the camera with two sensors. The corresponding paper is A robust registration method for UAV thermal infrared and visible images taken by dual-cameras.
We have released the full source code.
These implementation of methods (SIFT, SURF, ORB, RIFT, RCB, TFeat, HardNet,TWMM) have been released. One can run these methodsas follows:
git clone https://github.com/mlxljz/TWMM.git
python results_SIFT_SURF_RIFT_SCB_HOPC.py
The implementation for RIFT and SCB refers to http://www.ivlab.org/publications.html and https://ljy-rs.github.io/web/
The experimental results for different methods are located in this directory：test_img.
The experimental results for different methods are located in this directory：test_img/out.
The subitems in experimental results: homo, matchpoints_img, mosaic, warp_thermal
This repo is currently maintained by Lingxuan Meng (mlxuan123@163.com).