Truncation Trick

Introduced by Marchesi in Megapixel Size Image Creation using Generative Adversarial Networks

The Truncation Trick is a latent sampling procedure for generative adversarial networks, where we sample $z$ from a truncated normal (where values which fall outside a range are resampled to fall inside that range). The original implementation was in Megapixel Size Image Creation with GAN. In BigGAN, the authors find this provides a boost to the Inception Score and FID.

Source: Megapixel Size Image Creation using Generative Adversarial Networks

Read Paper See Code

Papers

Paper	Code	Results	Date	Stars

Tasks

Task	Papers	Share
Image Generation	27	31.40%
Conditional Image Generation	13	15.12%
Super-Resolution	4	4.65%
Image-to-Image Translation	3	3.49%
Denoising	2	2.33%
GAN inversion	2	2.33%
Image Super-Resolution	2	2.33%
Object Classification	2	2.33%
Image Classification	2	2.33%

Usage Over Time

This feature is experimental; we are continuously improving our matching algorithm.

Components

Component	Type	Add Remove
🤖 No Components Found	You can add them if they exist; e.g. Mask R-CNN uses RoIAlign

Categories

Add Remove

Latent Variable Sampling