CVNets: High Performance Library for Computer Vision

TL;DR

CVNets is an open-source library that enhances training efficiency and performance for various computer vision tasks, supporting multiple data handling and network implementation features.

Contribution

It introduces a comprehensive, high-performance library with novel data sampling methods and optimized network implementations for computer vision.

Findings

Achieves comparable or superior performance to existing methods.

Supports a wide range of visual recognition tasks.

Provides tools for efficient data handling and transformations.

Abstract

We introduce CVNets, a high-performance open-source library for training deep neural networks for visual recognition tasks, including classification, detection, and segmentation. CVNets supports image and video understanding tools, including data loading, data transformations, novel data sampling methods, and implementations of several standard networks with similar or better performance than previous studies. Our source code is available at: \url{https://github.com/apple/ml-cvnets}.