# vsitzmann/siren

Official implementation of "Implicit Neural Representations with Periodic Activation Functions"

Repository: https://github.com/vsitzmann/siren
Canonical: https://ross.abutalabs.com/products/vsitzmann-siren
Language: Python
License: MIT
License Family: permissive
Last push: 2024-07-27T18:22:54+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2262, "days_push": 767, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1993, forks 272 (observed 2026-08-28T04:06:03.277942+00:00)

## What it is
Official PyTorch implementation of SIREN, a neural network architecture using periodic (sine) activation functions for implicit neural representations of images, audio, video, and signed distance functions. Includes experiment scripts, differential operators, and a Colab notebook for interactive exploration.

## Use cases
- fit a neural network to an image or audio signal
- learn implicit neural representations of shapes and point clouds
- solve Poisson's equation with a neural network
- reproduce SIREN paper experiments
- experiment with periodic activation function initialization

## When to choose
- you need to represent continuous signals like images, audio, or SDFs with a small neural network
- you want the reference implementation of the SIREN paper
- you're researching coordinate-based networks and activation function design

## When to avoid
- you need a production-ready, actively maintained library
- you want general-purpose computer vision or audio processing tools
- you need a framework for standard feedforward or convolutional networks

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing, audio-processing, simulation
- domain: deep-learning, computer-vision, graphics
- platform: python
- tags: implicit-neural-representations, siren, neurips-paper, signal-representation, research-code, research, gpu

## Member repositories
- vsitzmann/siren (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:03.277942+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 2026-08-30T03:02:36.665303+00:00, confidence not recorded.
  - readme: https://github.com/vsitzmann/siren (fetched 2026-08-28T04:06:03.277942+00:00, sha b839ea402665)
- Data as of 2026-08-30T08:39:29.467469+00:00.