# rykov8/ssd_keras

Port of Single Shot MultiBox Detector to Keras

Repository: https://github.com/rykov8/ssd_keras
Canonical: https://ross.abutalabs.com/products/rykov8-ssd_keras
Language: Jupyter Notebook
License: MIT
License Family: permissive
Topics: computer-vision, object-detection, deep-learning, keras-models
Archived: true
Last push: 2021-06-01T16:00:23+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3589, "days_push": 1919, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1083, forks 540 (observed 2026-08-28T04:03:31.094943+00:00)

## What it is
A Keras port of the SSD (Single Shot MultiBox Detector) object detection model, including pretrained weights ported from the original Caffe implementation. It provides notebooks for forward-pass inference and training on 300x300 models, plus utilities for testing on video or camera input.

## Use cases
- run object detection on images with a pretrained SSD model in Keras
- train an SSD300 detector on a custom dataset
- detect objects in live video or webcam streams
- port Caffe SSD weights to a TensorFlow/Keras workflow
- study or reproduce the SSD paper implementation

## When to choose
- you need a classic single-shot object detector in Keras/TensorFlow
- you want pretrained SSD300 weights without training from scratch
- you are reproducing the SSD paper or learning object detection internals

## When to avoid
- you need modern detectors like YOLO or DETR with active development
- you require recent TensorFlow/Keras 2.x+ compatibility out of the box
- you need production-supported object detection with regular updates

## 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, ssd, keras, tensorflow, pretrained-weights

## Member repositories
- rykov8/ssd_keras (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:31.094943+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 2026-08-30T06:51:12.237516+00:00, confidence not recorded.
  - readme: https://github.com/rykov8/ssd_keras (fetched 2026-08-28T04:03:31.094943+00:00, sha 41cd381be989)
- Data as of 2026-08-30T08:39:29.467469+00:00.