# ZFTurbo/Weighted-Boxes-Fusion

Set of methods to ensemble boxes from different object detection models, including implementation of "Weighted boxes fusion (WBF)" method.

Repository: https://github.com/ZFTurbo/Weighted-Boxes-Fusion
Canonical: https://ross.abutalabs.com/products/weighted-boxes-fusion
Language: Python
License: MIT
License Family: permissive
Topics: ensemble-prediction, object-detection, boxes
Last push: 2026-07-27T10:59:12+00:00

## Health v2 (maintenance only)
Score: 65/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 94, release rhythm 8, longevity 100
- inputs: {"age_days": 2502, "days_push": 37, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1827, forks 235 (observed 2026-08-28T04:05:41.542101+00:00)

## What it is
A Python library implementing several methods for ensembling bounding boxes from multiple object detection models, including Non-maximum Suppression, Soft-NMS, Non-maximum Weighted, and Weighted Boxes Fusion (WBF). It supports 2D and 3D boxes and model weighting.

## Use cases
- ensemble predictions from multiple object detection models
- replace NMS with weighted boxes fusion for better accuracy
- merge bounding boxes from several models into one output
- apply soft-NMS to detection results
- fuse 3D bounding boxes from detection models
- improve mAP in detection competitions by combining model outputs

## When to choose
- you have predictions from multiple object detection models and want to combine them
- you want better ensembling results than plain NMS
- you need a lightweight pip-installable utility for box fusion in Python

## When to avoid
- you need full object detection model training or inference
- you work outside Python or need GPU-accelerated fusion at scale
- your task is not bounding-box based detection

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, computer-vision, image-processing
- domain: computer-vision, machine-learning, deep-learning
- platform: python, cross-platform
- tags: object-detection, ensemble, weighted-boxes-fusion, nms, bounding-boxes

## Member repositories
- ZFTurbo/Weighted-Boxes-Fusion (main) score 65

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:41.542101+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 2026-08-30T03:19:24.974987+00:00, confidence not recorded.
  - readme: https://github.com/ZFTurbo/Weighted-Boxes-Fusion (fetched 2026-08-28T04:05:41.542101+00:00, sha 1bda61f6f2fc)
- Data as of 2026-08-30T08:39:29.467469+00:00.