OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Cfnet

Cfnet is a Matlab research implementation of end-to-end representation learning with correlation filters for visual object tracking, providing pretrained networks, tracking evaluation code and example training workflows.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Cfnet — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Cfnet when you need a research codebase for pretrained correlation-filter tracking or want to explore training such networks, and can work with its Matlab and GPU-based dependencies.

Who it is for

  • Computer-vision researchers evaluating object tracking on video sequences.
  • Developers training correlation-filter tracking networks with the documented Matlab toolchain.

Documented uses

  • Run a pretrained tracker on video sequences and evaluate tracking performance.
  • Prepare compatible video data and train a tracking network using the repository’s example experiments.
Before you choose

Requirements & limits

The repository is public and specifies an MIT license.

  • The documented workflows require a GPU, CUDA, cuDNN, Matlab and MatConvNet.
  • The README cites an older tested software stack and cautions that other configurations are not guaranteed.
  • Training requires compatible curated data or the referenced metadata files; the metadata file is listed as 6.7 GB.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up pretrained tracking evaluation

Prepare Cfnet’s documented tracking-only workflow and start its video evaluation entry point.

Before you start

  • GPU, CUDA, cuDNN, Matlab and MatConvNet.
  • A cloned Cfnet repository and downloaded pretrained networks.
  • At least one video sequence in the expected format.

Steps

  1. Download and unzip the pretrained networks into cfnet/pretrained.

  2. In cfnet/src/tracking/, remove .example from the tracking environment and startup files, then edit them as appropriate.

  3. Place a suitable video sequence in the expected data layout; the README also describes using its validation set.

  4. Start from an entry point under cfnet/src/tracking/run_*_evaluation.m.

Expected outcome

The tracking evaluation workflow is configured to run against a supplied video sequence.

Guide limitations

  • The README cites Matlab 2015, MatConvNet v1.0beta24, CUDA 8.0 and cuDNN 5.1 as its tested setup, while the tracking prerequisites mention CUDA 7.5; other configurations are not guaranteed.
  • The repository reports problems from some users with Matlab 2017.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
bertinetto/cfnet · Official homepage title · retained in its source language
How we review sourcesSuggest a correction