# harvitronix/five-video-classification-methods

Code that accompanies my blog post outlining five video classification methods in Keras and TensorFlow

Repository: https://github.com/harvitronix/five-video-classification-methods
Canonical: https://ross.abutalabs.com/products/five-video-classification-methods
Homepage: https://medium.com/@harvitronix/five-video-classification-methods-implemented-in-keras-and-tensorflow-99cad29cc0b5
Language: Python
License: MIT
License Family: permissive
Topics: machine-learning, deep-learning, classification, keras, tensorflow
Last push: 2023-03-22T21:39:09+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": 3458, "days_push": 1260, "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 1182, forks 459 (observed 2026-08-28T04:03:54.400547+00:00)

## What it is
Companion code for a blog post demonstrating five deep learning approaches to video action classification using Keras and TensorFlow on the UCF101 dataset. It includes data extraction scripts, feature extraction, and training code for CNN, CNN+LSTM, LRCN, MLP, and 3D convolutional models.

## Use cases
- classify human actions in videos with deep learning
- learn different video classification architectures in Keras
- train CNN-LSTM models on the UCF101 dataset
- compare 3D CNN vs frame-based video classification methods
- extract per-frame features from videos with a ConvNet
- reproduce video action recognition experiments

## When to choose
- learning video classification techniques with worked examples
- comparing temporal modeling approaches (RNN, LRCN, 3D conv) on a standard dataset
- starting a UCF101-based research or class project

## When to avoid
- production video classification systems needing modern architectures
- projects requiring recent TensorFlow/Keras versions
- needing a ready-made inference demo or pretrained models

## Facets
- artifact type: learning-resource
- maturity: maintenance
- function: machine-learning, deep-learning, video-processing, image-processing
- domain: machine-learning, deep-learning, computer-vision, tutorials
- platform: python, cross-platform
- tags: keras, tensorflow, video-classification, action-recognition, ucf101, cnn, rnn, lstm, 3d-convolution, educational-code, video, gpu, linux

## Member repositories
- harvitronix/five-video-classification-methods (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:54.400547+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 2026-08-30T06:24:50.357499+00:00, confidence not recorded.
  - readme: https://github.com/harvitronix/five-video-classification-methods (fetched 2026-08-28T04:03:54.400547+00:00, sha 9487424417bf)
  - homepage: https://medium.com/@harvitronix/five-video-classification-methods-implemented-in-keras-and-tensorflow-99cad29cc0b5 (fetched 2026-08-29T12:32:03.560253+00:00, sha 0f96a0b8a783)
- Data as of 2026-08-30T08:39:29.467469+00:00.