SimSAM: Simple Siamese Representations Based Semantic Affinity Matrix for Unsupervised Image Segmentation

Error message

The specified file temporary://fileYIPHgl could not be copied, because the destination directory is not properly configured. This may be caused by a problem with file or directory permissions. More information is available in the system log.
The specified file temporary://filecFt3Pw could not be copied, because the destination directory is not properly configured. This may be caused by a problem with file or directory permissions. More information is available in the system log.
The specified file temporary://fileOBSqH7 could not be copied, because the destination directory is not properly configured. This may be caused by a problem with file or directory permissions. More information is available in the system log.
The specified file temporary://filePKSqv6 could not be copied, because the destination directory is not properly configured. This may be caused by a problem with file or directory permissions. More information is available in the system log.
The specified file temporary://fileum17M4 could not be copied, because the destination directory is not properly configured. This may be caused by a problem with file or directory permissions. More information is available in the system log.
The specified file temporary://fileQErUaI could not be copied, because the destination directory is not properly configured. This may be caused by a problem with file or directory permissions. More information is available in the system log.
The specified file temporary://fileTtfzya could not be copied, because the destination directory is not properly configured. This may be caused by a problem with file or directory permissions. More information is available in the system log.
The specified file temporary://filelNlg7i could not be copied, because the destination directory is not properly configured. This may be caused by a problem with file or directory permissions. More information is available in the system log.
The specified file temporary://fileq3hnvu could not be copied, because the destination directory is not properly configured. This may be caused by a problem with file or directory permissions. More information is available in the system log.

Citation Author(s):: Chanda Grover Kamra, Indra Deep Mastan, Debayan Gupta
Submitted by:: Chanda Grover Kamra
Last updated:: 8 February 2024 - 3:06am
Document Type:: Supplementary Material
Document Year:: 2024
Presenters:: Chanda Grover Kamra
Paper Code:: 1183

Categories:: Image/Video Processing
Image/Video Processing

Recent developments in self-supervised learning (SSL) have made it possible to learn data representations without the need for annotations.
, which enhances the performance of various downstream tasks.
Inspired by the non-contrastive SSL approach (SimSiam), we introduce a novel framework SimSAM to compute the Semantic Affinity Matrix, which is significant for unsupervised image segmentation. Given an image, SimSAM first extracts features using pre-trained DINO-ViT, then projects the features to predict the correlations of dense features in a non-contrastive way. We show applications of the Semantic Affinity Matrix in object segmentation and semantic segmentation tasks. Our code is available at https://anonymous.4open.science/r/SimSAM-E746/.