DreamFactor

Improve ML model performance and reduce bias with synthetic data

Case

1000x cost saving for LEGO part sorting


400k labeled class instances generated in a few hours to train a robust object detection model for LEGO parts.

Download the dataset >

Whitepaper

Improved model performance on PASCAL VOC


Mixing DreamFactor synthetic data with baseline dataset increases mAP.

Whitepaper coming soon.

How it works


1. Upload 3D models

Upload 3D models of the objects you want to detect.

2. Set generation parameters

Control resolution, render quality, lighting and other parameters.

3. Download dataset

Download images and labels for use with PyTorch, TensorFlow and others.

Private beta

Generate a synthetic dataset

Industries

Warehousing

Increase density and speed of delivery with improved irregular product perception.

Delivery robots

Decrease delivery time by improving perception of anomalous events.

Retail

Deeply understand consumer behavior and decision making around specific products.

Autonomous vehicles

Improve safety and comfort with robust training for long-tail events.

Private beta

Generate a synthetic dataset

Synthetic data generation is a cost-effective and flexible way to get perfectly labeled image data for object detection and classification machine learning tasks.

© NewNoetic, Inc. 2021. All rights reserved.

Private beta

Generate a synthetic dataset

DreamFactor

Improve image classification model performance
and reduce bias with synthetic data

Accumsan

  1. Lorem ipsum dolor sit amet, consectetur elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.

  1. Lorem ipsum dolor sit amet, consectetur elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.

Nulla fermentum

Lorem ipsum dolor sit amet, consectetur ut labore et dolore magna aliqua. Ut enim ad minim nostrud sed ullamco.