EmerDiff: Emerging Pixel-level Semantic Knowledge in Diffusion Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Namekata, Koichi, Sabour, Amirmojtaba, Fidler, Sanja, Kim, Seung Wook
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913203188924416
author Namekata, Koichi
Sabour, Amirmojtaba
Fidler, Sanja
Kim, Seung Wook
author_facet Namekata, Koichi
Sabour, Amirmojtaba
Fidler, Sanja
Kim, Seung Wook
contents Diffusion models have recently received increasing research attention for their remarkable transfer abilities in semantic segmentation tasks. However, generating fine-grained segmentation masks with diffusion models often requires additional training on annotated datasets, leaving it unclear to what extent pre-trained diffusion models alone understand the semantic relations of their generated images. To address this question, we leverage the semantic knowledge extracted from Stable Diffusion (SD) and aim to develop an image segmentor capable of generating fine-grained segmentation maps without any additional training. The primary difficulty stems from the fact that semantically meaningful feature maps typically exist only in the spatially lower-dimensional layers, which poses a challenge in directly extracting pixel-level semantic relations from these feature maps. To overcome this issue, our framework identifies semantic correspondences between image pixels and spatial locations of low-dimensional feature maps by exploiting SD's generation process and utilizes them for constructing image-resolution segmentation maps. In extensive experiments, the produced segmentation maps are demonstrated to be well delineated and capture detailed parts of the images, indicating the existence of highly accurate pixel-level semantic knowledge in diffusion models.
format Preprint
id arxiv_https___arxiv_org_abs_2401_11739
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle EmerDiff: Emerging Pixel-level Semantic Knowledge in Diffusion Models
Namekata, Koichi
Sabour, Amirmojtaba
Fidler, Sanja
Kim, Seung Wook
Computer Vision and Pattern Recognition
Machine Learning
Diffusion models have recently received increasing research attention for their remarkable transfer abilities in semantic segmentation tasks. However, generating fine-grained segmentation masks with diffusion models often requires additional training on annotated datasets, leaving it unclear to what extent pre-trained diffusion models alone understand the semantic relations of their generated images. To address this question, we leverage the semantic knowledge extracted from Stable Diffusion (SD) and aim to develop an image segmentor capable of generating fine-grained segmentation maps without any additional training. The primary difficulty stems from the fact that semantically meaningful feature maps typically exist only in the spatially lower-dimensional layers, which poses a challenge in directly extracting pixel-level semantic relations from these feature maps. To overcome this issue, our framework identifies semantic correspondences between image pixels and spatial locations of low-dimensional feature maps by exploiting SD's generation process and utilizes them for constructing image-resolution segmentation maps. In extensive experiments, the produced segmentation maps are demonstrated to be well delineated and capture detailed parts of the images, indicating the existence of highly accurate pixel-level semantic knowledge in diffusion models.
title EmerDiff: Emerging Pixel-level Semantic Knowledge in Diffusion Models
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2401.11739