OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Image & Video AI

Tfjs Yolo Tiny

Tfjs Yolo Tiny is a JavaScript library that runs Tiny YOLO object detection in the browser using TensorFlow.js and returns detected classes, probabilities and bounding-box coordinates.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Tfjs Yolo Tiny — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when a browser application needs Tiny YOLO detections from images or frames and can work with the library’s TensorFlow.js model and image-tensor interface.

Who it is for

  • JavaScript developers adding object detection to browser-based image workflows.
  • Developers prototyping detection on webcam captures or video frames.

Documented uses

  • Detect and label objects in an image tensor in a web application.
  • Retrieve bounding-box coordinates and class probabilities for drawing overlays.
  • Adjust detection filtering or configure supported custom Tiny YOLO models.
Before you choose

Requirements & limits

  • The package is installed through npm or Yarn and exposes JavaScript functions for model loading and detection.
  • The documented default input is an RGB tensor of shape (1, 416, 416, 3).
  • The README reports about 800 ms per frame on one older MacBook Pro and Chrome setup, so its stated video pace is roughly one frame per second.
  • Custom model options include input dimensions, class count, class names and YOLO anchors; anchors are noted for retraining on a new dataset.
  • The repository specifies an MIT license.

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

Add Tiny YOLO detection to a JavaScript app

Install the package, load a Tiny YOLO model, and pass an image tensor to the detector.

Before you start

  • A JavaScript project using TensorFlow.js
  • An RGB image tensor shaped (1, 416, 416, 3)
  • A Tiny YOLO TensorFlow.js model

Steps

  1. Install the package with the command below or yarn add tfjs-yolo-tiny.

    Documented command
    npm install tfjs-yolo-tiny

  2. Import yolo and downloadModel from tfjs-yolo-tiny.

  3. Load the model with await downloadModel().

  4. Call await yolo(inputImage, model) with the image tensor and loaded model.

  5. Use each returned box’s coordinates, class name and class probability in the application.

Expected outcome

An array of detected objects with bounding-box coordinates, class names and class probabilities, ready for application-side display.

Guide limitations

  • The documented default input is an RGB tensor shaped (1, 416, 416, 3).
  • The implementation supports Tiny YOLO; the README says the referenced TensorFlow.js setup did not support full YOLO models.
  • The README reports approximately 800 ms per frame on Chrome on a 13-inch, mid-2014 MacBook Pro; other performance is not specified.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
ModelDepot/tfjs-yolo-tiny · Official homepage title · retained in its source language
How we review sourcesSuggest a correction