Can I read Unsupervised Visual Tracking Method on EtoBox?

Unsupervised Visual Tracking Method by Luigy Machaca is a document available to read on EtoBox.

What is Unsupervised Visual Tracking Method about?

This document proposes an unsupervised deep tracking method that trains a CNN model on unlabeled video data by exploiting forward and backward target predictions. The model is trained to track the target object forward through successive frames and backward to the initial frame. This is done using a Siamese correlation filter network on unlabeled videos. The method aims to learn generic representations without requiring annotated data.

Author

Luigy Machaca

Language

EN