Segment scene using DINO-ViT feature space and predict embedding that preserve semantics with SimSAM
Segment scene using DINO-ViT feature space and predict embedding that preserve semantics with SimSAM
SimSAM: Simple Siamese Representations Based Semantic Affinity Matrix for Unsupervised Image Segmentation
arXiv paper abstract https://arxiv.org/abs/2406.07986
arXiv PDF paper https://arxiv.org/pdf/2406.07986
GitHub https://github.com/chandagrover/SimSAM
Recent developments in self-supervised learning (SSL) have made it possible to learn data representations without the need for annotations.
Inspired by the non-contrastive SSL approach (SimSiam), … 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.
… show applications of the Semantic Affinity Matrix in object segmentation and semantic segmentation tasks …
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website