HaiPenglai/

Succulent-Vision-Dataset

	Succulent Vision Dataset




	
	Introduction

This dataset contains 1,000+ high-quality segmented images of succulents. The original photos were taken in complex cluster environments, and individual succulent plants were precisely segmented using SAM (Segment Anything Model).

To provide structured data, the images have been automatically categorized using an advanced unsupervised pipeline:

Feature Extraction: DINOv2 (Vision Transformer) for… See the full description on the dataset page: https://huggingface.co/datasets/HaiPenglai/Succulent-Vision-Dataset.