Home › Categories › Technology› TV-L1 Flow

TV-L1 Flow.pdf

article_lr.pdf

🔗 Source: ipol.im

📊 Size: 662 KB

👤 Author: Javier Sánchez, Enric Meinhardt-Llopis, Gabriele Facciolo

⬇️ Downloads: 105

Summary

TV-L1 optical flow estimation is based on minimizing a functional with a data term using the L1 norm and a regularization term using the total variation of the flow, allowing discontinuities in the flow field and being more robust to noise than the classical Horn and Schunck approach. The algorithm solves a relaxed version of the problem by alternate minimization, with a C implementation and online demo available. The method uses a convex relaxation and can be minimized by fixing one variable and solving for the other, with the first sub-problem fitting the total variation denoising model and the second sub-problem solved point-wise by thresholding. The numerical scheme uses a fixed-point iteration and thresholding operation, with the input being a pair of images and the output being a vector field.

Download File

Description

TV-L1 optical flow estimation is based on minimizing a functional with a data term using the L1 norm and a regularization term using the total variation of the...

Technical Information

File Format: PDF
File Size: 662 KB
Pages: 14
Language: EN
Author: Javier Sánchez, Enric Meinhardt-Llopis, Gabriele Facciolo
Total Downloads: 105
Last Updated: 1 week ago

Document Overview

This PDF document about TV-L1 Flow provides comprehensive information and guidance. Whether you're a beginner or advanced user, this resource offers valuable insights into TV-L1 Flow.