# Duankaiwen/CenterNet

Codes for our paper "CenterNet: Keypoint Triplets for Object Detection" .

Repository: https://github.com/Duankaiwen/CenterNet
Canonical: https://ross.abutalabs.com/products/duankaiwen-centernet
Language: Python
License: MIT
License Family: permissive
Topics: object-detection, one-stage, keypoint-triplets
Last push: 2022-04-18T15:22:27+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2696, "days_push": 1598, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1884, forks 371 (observed 2026-08-28T04:05:48.924037+00:00)

## What it is
Official PyTorch implementation of CenterNet, a one-stage keypoint-triplet-based object detector built on CornerNet. It trains from scratch and achieves 47.0% AP on MS-COCO, surpassing prior one-stage detectors.

## Use cases
- train an object detection model on MS-COCO
- evaluate a keypoint-based one-stage detector
- reproduce CenterNet paper results
- compare one-stage vs two-stage detector performance
- build object detection research on keypoint triplets
- use cascade corner pooling and center pooling modules

## When to choose
- you need a strong one-stage object detector for research or benchmarking
- you want to reproduce or extend the CenterNet paper
- you study keypoint-based detection methods like CornerNet

## When to avoid
- you need a production-ready, actively maintained detection framework
- you want plug-and-play pretrained detectors with modern tooling
- you need the latest state-of-the-art architectures

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision
- domain: computer-vision, deep-learning, machine-learning
- platform: python
- tags: object-detection, keypoint-detection, one-stage-detector, pytorch, research-code, ms-coco, linux, gpu

## Member repositories
- Duankaiwen/CenterNet (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:48.924037+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:13:37.943861+00:00, confidence not recorded.
  - readme: https://github.com/Duankaiwen/CenterNet (fetched 2026-08-28T04:05:48.924037+00:00, sha f4dc45c0696c)
- Data as of 2026-08-30T08:39:29.467469+00:00.