OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Fishial Fish Identification

Fishial Fish Identification is a public, MIT-licensed repository of training and validation scripts, models and helper notebooks for fish image segmentation, object detection and classification. It also includes a browser-based 3D viewer for exploring classification embeddings across 866 fish species.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Fishial Fish Identification — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you need a code-based starting point for training, adapting or exploring fish image-recognition models, rather than only a finished image-identification interface.

Who it is for

  • Developers adapting or training fish-recognition models.
  • Researchers exploring fish-species image classification and embeddings.
  • Teams with labelled fish images preparing segmentation, classification or detection datasets.

Documented uses

  • Train a fish segmentation model from COCO annotations and images.
  • Train a fish image classifier from image directories and annotations.
  • Run a YOLO-based object-detection training workflow with dataset configuration.
  • Explore species embeddings, nearest neighbours and similarity comparisons in the 3D viewer.
  • Prepare classification data or create embeddings and TorchScript exports using the included helpers.
Before you choose

Requirements & limits

The repository is public and identifies its license as MIT. Its README links to a Google Colab notebook and documents local setup with Python 3.10 or newer.

  • The main resource is a repository of scripts and helpers, so training workflows require appropriately prepared datasets and annotations.
  • The README distinguishes Detectron2 segmentation training from YOLO object detection and classification workflows.
  • The repository documents a Colab route as well as local setup; local requirements specify Python 3.10 or newer and pinned dependencies.
  • The embedding viewer is a separate browser-based tool featuring 866 species, nearest-neighbour exploration and comparison of up to five species.
  • TorchScript export helpers are documented for PyTorch classification and Detectron2 segmentation models; supported inference runtimes and an inference API are not specified.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up the repository locally

Prepare the documented Python environment for Fishial’s training scripts.

Before you start

  • Git and Python 3.10 or newer.
  • The repository’s requirements.txt file and pinned dependencies.

Steps

  1. Clone the repository and change into its directory.

  2. Install dependencies with pip3 install -r requirements.txt.

  3. Check that python3 --version reports Python 3.10 or newer.

Expected outcome

A local Python environment prepared according to the repository’s setup instructions.

Guide limitations

  • This covers environment setup only; it does not run training or verify compatibility with a particular machine.
  • The README advises recreating the virtual environment if the Python version changes.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
Wye-Foundation/Fishial-Fish-Identification · Official homepage title · retained in its source language
How we review sourcesSuggest a correction