wongkinyiu/yolov7

★ 14,092Jupyter NotebookAudience · researcherComplexity · 4/5Setup · moderate

Mindmap

mindmap
  root((YOLOv7))
    What it does
      Object detection
      Real-time video
      161 fps on one GPU
    Model variants
      Standard YOLOv7
      Larger E6E variant
      Speed vs accuracy
    Capabilities
      Pose estimation
      Instance segmentation
      Pre-trained weights
    Training
      Single GPU script
      Multi-GPU script
      Transfer learning
    Setup
      Docker recommended
      pip alternative
      Hugging Face demo

mindmap root((YOLOv7)) What it does Object detection Real-time video 161 fps on one GPU Model variants Standard YOLOv7 Larger E6E variant Speed vs accuracy Capabilities Pose estimation Instance segmentation Pre-trained weights Training Single GPU script Multi-GPU script Transfer learning Setup Docker recommended pip alternative Hugging Face demo

Click or tap to explore — scroll the page freely

Things people build with this

USE CASE 1

Run real-time object detection on a live video feed to identify people, cars, or animals without training anything.

USE CASE 2

Fine-tune YOLOv7 on your own labeled image dataset to detect custom objects specific to your use case.

USE CASE 3

Use the pose estimation feature to detect and overlay human body keypoints on images or video.

USE CASE 4

Apply instance segmentation to draw exact outlines around each detected object rather than just bounding boxes.

Tech stack

PythonPyTorchCUDAJupyter NotebookDocker

Getting it running

Difficulty · moderate Time to first run · 30min

Docker is recommended for setup, a CUDA GPU is needed for training and speeds up inference significantly.

In plain English

YOLOv7 is an object detection system that can identify what is in an image or video frame and mark each recognized object with a bounding box, all in real time. Object detection means automatically finding and labeling things like people, cars, or animals within a photo or video. This repository contains the official research code from a 2022 paper that introduced the YOLOv7 architecture. The model runs on PyTorch and is built around speed. The base YOLOv7 variant processes images at 161 frames per second on a single GPU, fast enough for live video feeds. Several size variants are included, from the standard model to larger ones like YOLOv7-E6E that trade some speed for better accuracy. All variants come with pre-trained weight files you can download and use immediately without training anything yourself. If you want to train on your own images, the repository provides scripts for single-GPU and multi-GPU setups. You point the training script at a folder of labeled images, set a configuration file, and run it. Transfer learning is supported too: you can start from one of the provided checkpoints rather than from scratch, which is useful when your dataset is small. The code also includes support for pose estimation (detecting human body keypoints in images) and instance segmentation (drawing the exact outline of each detected object rather than just a box around it). A live web demo is hosted on Hugging Face Spaces if you want to try the model without any local setup. The recommended installation route is Docker, with pip as an alternative for environments where Docker is not practical.

Copy-paste prompts

Prompt 1

Using YOLOv7, write a Python script that loads the pre-trained weights and runs object detection on a folder of images, saving annotated output images.

Prompt 2

How do I fine-tune YOLOv7 on a custom dataset of 500 labeled images using transfer learning from the pretrained YOLOv7 checkpoint?

Prompt 3

Set up YOLOv7 with Docker and run real-time object detection on my webcam feed to detect people.

Prompt 4

How do I use YOLOv7's pose estimation model to detect human body keypoints in a video file and overlay the skeleton on each frame?

Prompt 5

What is the speed vs accuracy trade-off between YOLOv7 and YOLOv7-E6E, and when should I use each one?

Open on GitHub → Explain another repo

← wongkinyiu on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.