Ajou University repository

Entropy-weighted Voting Method for Diffusion-based Semantic Segmentation
Citations

SCOPUS

0

Citation Export

Publication Year
2023-01-01
Journal
International Conference on ICT Convergence
Publisher
IEEE Computer Society
Citation
International Conference on ICT Convergence, pp.304-307
Keyword
diffusionentropysemantic segmentationvoting
Mesh Keyword
Diffusion modelDiffusion networksImage generationsMultilayers perceptronsPerformanceSegmentation methodsSemantic segmentationSimple majorityVotingWeighted voting methods
All Science Classification Codes (ASJC)
Information SystemsComputer Networks and Communications
Abstract
Recently, semantic segmentation methods leveraging image generation models have garnered significant attention. In particular, approach based on Diffusion models (DDPM) that utilize mid-level activations from the diffusion network with a majority voting of distributions from several light multi-layer perceptron (MLP) have shown better performance compared to GAN-based approaches. However, utilizing a simple majority voting system is suboptimal. In this paper, we propose a novel voting method for DDPM-based semantic segmentation. Our method introduces a weighted sum of distributions, where the weights are determined by the entropy of the class prediction results obtained from each MLP model. We conduct experiments on various datasets, including LSUN-Bedroom, FFHQ-256, LSUN-Cat, and LSUN-Horse. The results demonstrate that our proposed method achieves better mean Intersection over Union (mIoU) scores compared to previous work.
Language
eng
URI
https://aurora.ajou.ac.kr/handle/2018.oak/36963
https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85184614264&origin=inward
DOI
https://doi.org/10.1109/ictc58733.2023.10393545
Journal URL
http://ieeexplore.ieee.org/xpl/conferences.jsp
Type
Conference
Funding
This work has been supported by the MSIT(Ministry of Science and ICT), Korea, under the ITRC(Information Technology Research Center) support program(IITP-2023-2018-0-01424) supervised by the IITP(Institute for Information communications Technology Promotion).
Show full item record

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

Heo,Yong Seok  Image
Heo,Yong Seok 허용석
Department of Electrical and Computer Engineering
Read More

Total Views & Downloads

File Download

  • There are no files associated with this item.