# facebookresearch/video-nonlocal-net

Non-local Neural Networks for Video Classification

Repository: https://github.com/facebookresearch/video-nonlocal-net
Canonical: https://ross.abutalabs.com/products/video-nonlocal-net
Language: Python
License: NOASSERTION
License Family: other
Archived: true
Last push: 2021-09-15T17:26:10+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": 3141, "days_push": 1813, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1990, forks 320 (observed 2026-08-28T04:06:03.042294+00:00)

## What it is
A research codebase from Facebook AI implementing Non-local Neural Networks for video classification, built on the Caffe2 framework. It reproduces the CVPR 2018 paper experiments on the Kinetics dataset with pre-trained models and training scripts.

## Use cases
- reproduce non-local neural network video classification results
- train video classification models on Kinetics
- download pre-trained non-local network models
- study non-local operations in deep learning
- compare baseline vs non-local video model benchmarks

## When to choose
- you need the exact reference implementation of the Non-local Neural Networks paper
- you are doing research on non-local attention blocks for video
- you can work with the legacy Caffe2 framework

## When to avoid
- you need a maintained library or modern framework support (PyTorch/TensorFlow)
- you want production video classification
- you cannot use the deprecated Caffe2 dependency
- you need a permissive license (code is CC-BY-NC 4.0)

## Facets
- artifact type: library
- maturity: abandoned
- function: deep-learning, machine-learning, video-processing, image-processing
- domain: deep-learning, computer-vision, machine-learning
- platform: python
- tags: non-local-neural-networks, video-classification, caffe2, kinetics-dataset, research-code, cvpr-2018, video, linux, gpu

## Member repositories
- facebookresearch/video-nonlocal-net (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:03.042294+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 2026-08-30T03:02:43.834705+00:00, confidence not recorded.
  - readme: https://github.com/facebookresearch/video-nonlocal-net (fetched 2026-08-28T04:06:03.042294+00:00, sha c09d9c1bb30c)
- Data as of 2026-08-30T08:39:29.467469+00:00.