mychina75 / FTMU

Fast Template Matching and Update for Video Object Tracking and Segmentation

Prerequisites

• Python 3.7.5
• Pytorch 1.3.1
• torchvision 0.4.2
• CUDA 10.1

You can download the pre-trained model here, the training log file here, and the segmentation results here.

Please modify the dataset_root in parameter.py for the dataset path. The model is evaluated after each epoch, with the accuracy printed. The log file is saved in ./log, and the segmentation results are saved in ./best_results.

python ./train_eval.py

You can also see the original AC training code in AC_train.py

python ./AC_train.py