Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jul 16, 2026

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

Published on: January 18, 2020

Adapting a Foundation Monocular Depth Model for Soccer Video: From Synthetic Supervision to Match-Level Reliability.

Ju-Seong Do¹, Ho-Young Jung¹

¹Department of Artificial Intelligence, Kyungpook National University, 80 Daehak-ro, Buk-gu, Daegu 41566, Republic of Korea.

Sensors (Basel, Switzerland)

|July 15, 2026

Summary

Related Concept Videos

Depth Perception and Spatial Vision

Depth Perception and Spatial Vision

Depth perception is the ability to perceive objects three-dimensionally. It relies on two types of cues: binocular and monocular. Binocular cues depend on the combination of images from both eyes and how the eyes work together. Since the eyes are in slightly different positions, each eye captures a slightly different image. This disparity between images, known as binocular disparity, helps the brain interpret depth. When the brain compares these images, it determines the distance to an object.

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

A disambiguation framework for refining and answering ambiguous questions.

Scientific reports·2026

Same author

Improved Spoken Language Representation for Intent Understanding in a Task-Oriented Dialogue System.

Sensors (Basel, Switzerland)·2022

See all related articles

This study adapted a Depth Anything V2 model for soccer video analysis, improving depth estimation accuracy and reliability across matches. The adapted model offers a viable solution for post-match analysis, enhancing scene understanding beyond the pitch plane.

Area of Science:

Computer Vision
Machine Learning
Sports Analytics

Background:

Soccer video analysis traditionally focuses on pitch-plane tracking.
Camera-view depth cues like occlusion and goal structure are underrepresented.
Synthetic benchmarks offer dense supervision but real-world adaptation feasibility is unclear.

Purpose of the Study:

Evaluate a Depth Anything V2 model adapted to SoccerNet-Depth.
Assess unaligned MDE accuracy, scale-and-shift alignment, match-to-match reliability, and accuracy-cost trade-off.
Determine operational feasibility for soccer video depth estimation.

Main Methods:

Adapted the Depth Anything V2 model using the SoccerNet-Depth dataset.
Evaluated performance using unaligned and aligned metrics (AbsRel).

Keywords:

Depth Anything V2 computational efficiency foundation model adaptation match-level reliability monocular depth estimation soccer-video analysis sports computer vision

Related Experiment Videos

Last Updated: Jul 16, 2026

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

Published on: January 18, 2020

Conducted match-to-match reliability tests and accuracy-cost analysis.

Main Results:

Achieved an unaligned validation AbsRel of 0.00372.
Adaptation improved all eight metrics across 21 validation matches compared to a reference.
Reduced AbsRel by 34.1% on the challenge split versus the official baseline.
Higher resolution improved AbsRel by 5.9%, but default offered better accuracy-cost balance.

Conclusions:

The adapted Depth Anything V2 model demonstrates significant improvements in depth estimation for soccer videos.
The adaptation protocol provides a benchmark-scoped case study for foundation MDE on SoccerNet-Depth.
The default model configuration is suitable for post-match analysis due to its accuracy-cost balance.