Segment object using stable diffusion to fine-tune SAM with ASAM
Segment object using stable diffusion to fine-tune SAM with ASAM
ASAM: Boosting Segment Anything Model with Adversarial Tuning
arXiv paper abstract https://arxiv.org/abs/2405.00256
arXiv PDF paper https://arxiv.org/pdf/2405.00256
GitHub https://github.com/luckybird1994/ASAM
Hugging Face https://huggingface.co/spaces/xhk/ASAM
Project page https://asam2024.github.io
… Segment Anything Model (SAM) by Meta AI has distinguished itself in image segmentation.
… This paper introduces ASAM, a novel methodology that amplifies SAM’s performance through adversarial tuning.
… By utilizing a stable diffusion model, … augment a subset (1%) of the SA-1B dataset, generating adversarial instances that are more representative of natural variations rather than conventional imperceptible perturbations.
… approach maintains the photorealism of adversarial examples and ensures alignment with original mask annotations, thereby preserving the integrity of the segmentation task.
The fine-tuned ASAM demonstrates significant improvements across a diverse range of segmentation tasks without necessitating additional data or architectural modifications.
… evaluations confirm that ASAM establishes new benchmarks in segmentation tasks, thereby contributing to the advancement of foundational models in computer vision …
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
LinkedIn https://www.linkedin.com/in/morris-lee-47877b7b
