# lufficc/SSD

High quality, fast, modular reference implementation of SSD in PyTorch

Repository: https://github.com/lufficc/SSD
Canonical: https://ross.abutalabs.com/products/ssd
Language: Python
License: MIT
License Family: permissive
Topics: computer-vision, deep-learning, object-detection, pytorch, ssd
Last push: 2024-01-26T01:50:32+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2827, "days_push": 951, "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 1585, forks 384 (observed 2026-08-28T04:05:07.600334+00:00)

## What it is
A high-quality, fast, modular reference implementation of the SSD (Single Shot MultiBox Detector) object detection model in PyTorch. It supports multi-GPU training and inference, modular component swapping (backbones, heads), CPU inference, and evaluation with metrics visualization via TensorBoard.

## Use cases
- train an SSD object detection model on Pascal VOC or COCO
- run object detection inference on images with a pretrained SSD model
- research new SSD variants by swapping backbones or detection heads
- benchmark object detection training speed in PyTorch
- fine-tune object detection on a custom dataset
- visualize training metrics like mAP and AP per category

## When to choose
- you need a clean, modular PyTorch codebase for SSD-based object detection research
- you want multi-GPU training with DistributedDataParallel out of the box
- you want to easily replace components like the backbone (e.g., EfficientNet) via config
- you need CPU inference support and automatic pre-trained weight downloads

## When to avoid
- you need state-of-the-art detectors like YOLO, DETR, or Faster R-CNN rather than SSD
- you want a production-ready inference service rather than a research codebase
- you need support for frameworks other than PyTorch
- you require active development of new features, as the project is in maintenance mode

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision, image-processing
- domain: computer-vision, deep-learning, machine-learning, image-processing
- platform: python, windows
- tags: object-detection, ssd, pytorch, single-shot-detector, reference-implementation, multi-gpu-training, pascal-voc, coco, linux, macos, gpu

## Member repositories
- lufficc/SSD (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:07.600334+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 2026-08-30T03:55:37.546189+00:00, confidence not recorded.
  - readme: https://github.com/lufficc/SSD (fetched 2026-08-28T04:05:07.600334+00:00, sha a25514b54280)
- Data as of 2026-08-30T08:39:29.467469+00:00.