# Generic Perceptual Loss for Modeling Structured Output Dependencies

Source: [Generic Perceptual Loss for Modeling Structured Output Dependencies](<https://www.sota2.com/research/paper/68fa69971e209131cea22e50>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-04T05:05:18.911Z

Authors: Yifan Liu, Hao Chen, Yu Chen, Wei Yin, Chunhua Shen

Published: 2021-03-18T23:56:07.000Z

## Abstract

The perceptual loss has been widely used as an effective loss term in image synthesis tasks including image super-resolution, and style transfer. It was believed that the success lies in the high-level perceptual feature representations extracted from CNNs pretrained with a large set of images. Here we reveal that, what matters is the network structure instead of the trained weights. Without any learning, the structure of a deep network is sufficient to capture the dependencies between multiple levels of variable statistics using multiple layers of CNNs. This insight removes the requirements of pre-training and a particular network structure (commonly, VGG) that are previously assumed for the perceptual loss, thus enabling a significantly wider range of applications. To this end, we demonstrate that a randomly-weighted deep CNN can be used to model the structured dependencies of outputs. On a few dense per-pixel prediction tasks such as semantic segmentation, depth estimation and instance segmentation, we show improved results of using the extended randomized perceptual loss, compared to the baselines using pixel-wise loss alone. We hope that this simple, extended perceptual loss may serve as a generic structured-output loss that is applicable to most structured output learning tasks.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2103.10571>)

[PDF](<https://arxiv.org/pdf/2103.10571.pdf>)

[Implementation](<https://git.io/AdelaiDet>)
