Computer Vision Python Packages

datasets

🤗 The largest hub of ready-to-use datasets for AI models with fast, easy-to-use and efficient data manipulation tools

scikit-image

Image processing in Python

torchvision

Datasets, Transforms and Models specific to Computer Vision

ultralytics

Ultralytics YOLO27, YOLO26, YOLO11, YOLOv8 — object detection, instance segmentation, semantic segmentation, image classification, pose estimation, object tracking

ultralytics-thop

PyTorch model profiler for computing MACs and parameter counts to measure deep learning model complexity.

kornia

🐍 Geometric Computer Vision Library for Spatial AI

mediapipe

Cross-platform, customizable ML solutions for live and streaming media.

kornia-rs

🦀 Low-level 3D Computer Vision library in Rust

pytorch-metric-learning

The easiest way to use deep metric learning in your application. Modular, flexible, and extensible. Written in PyTorch.

open-clip-torch

An open source implementation of CLIP.

ultralytics-platform

Official Python SDK for the Ultralytics Platform API

rerun-sdk

Visualize, query, and stream to train on multimodal robotics data.

openvino

OpenVINO™ is an open source toolkit for optimizing and deploying AI inference

supervision

We write your reusable computer vision tools. 💜

segmentation-models-pytorch

Semantic segmentation models with 500+ pretrained convolutional and transformer-based backbones.

faster-coco-eval

Continuation of an abandoned project fast-coco-eval

roboflow

The official Roboflow Python package. Manage your datasets, models, and deployments. Roboflow has everything you need to build a computer vision application.

pycolmap

COLMAP - Structure-from-Motion and Multi-View Stereo

autogluon-core

Fast and Accurate ML in 3 Lines of Code

autogluon-features

Fast and Accurate ML in 3 Lines of Code

autogluon-tabular

Fast and Accurate ML in 3 Lines of Code

autogluon-common

Fast and Accurate ML in 3 Lines of Code

mmengine

OpenMMLab Foundational Library for Training Deep Learning Models

rfdetr

RF-DETR is a real-time object detection and segmentation model architecture developed by Roboflow, SOTA on COCO, designed for fine-tuning. [ICLR 2026]