Open-source sam2-hiera-base-plus Mask Generation Model - Free to Use Meta's Advanced SAM2 Base Variant

Sam2 Hiera Base Plus

Developed by merve

SAM2 is the state-of-the-art mask generation model released by Meta. This repository contains its base variant.

Image Segmentation Open Source License:Apache-2.0 #High-precision mask generation #Zero-shot segmentation #Video object segmentation

Downloads 20

Release Time : 8/2/2024

Model Overview

SAM2 is an advanced model for image segmentation, capable of generating high-quality masks and supporting mask generation via prompts or automatically.

Model Features

High-quality mask generation

Capable of generating precise image segmentation masks suitable for various complex scenarios.

Supports prompt and automatic generation

Masks can be generated either via prompts (e.g., bounding boxes) or automatically for all objects in the image.

Suitable for video object segmentation

Extended support for video object segmentation tasks, ideal for dynamic scene analysis.

Model Capabilities

Image segmentation

Mask generation

Video object segmentation

Use Cases

Computer vision

Image segmentation

Accurately segment objects in images to generate high-quality masks.

High-precision segmentation results suitable for complex scenes.

Video object segmentation

Track and segment objects in videos, ideal for dynamic scene analysis.

Stable object tracking and segmentation performance.

🚀 SAM2-Hiera-base-plus

This repository houses the base variant of the SAM2 model, a state-of-the-art mask generation model released by Meta.

🚀 Quick Start

This repository contains the base variant of the SAM2 model. SAM2 is the state-of-the-art mask generation model released by Meta.

✨ Features

Mask Generation: Capable of generating masks for various tasks.
Multiple Usage Modes: Supports both manual prompting and automatic mask generation.

📦 Installation

You can install the packaged version of SAM2 as follows:

pip install samv2 huggingface_hub

💻 Usage Examples

Basic Usage

from huggingface_hub import hf_hub_download
from sam2.build_sam import build_sam2
from sam2.sam2_image_predictor import SAM2ImagePredictor

hf_hub_download(repo_id = "merve/sam2-hiera-base-plus", filename="sam2_hiera_base_plus.pt", local_dir = "./")
sam2_checkpoint = "../checkpoints/sam2_hiera_base_plus.pt"
model_cfg = "sam2_hiera_b+.yaml"

sam2_model = build_sam2(config, ckpt, device="cuda", apply_postprocessing=False)
predictor = SAM2ImagePredictor(sam2_model)

# it accepts coco format
box = [x1, y1, w, h]
predictor.set_image(image)

masks = predictor.predict(box=box,
multimask_output=False)

Advanced Usage

For automatic mask generation:

from huggingface_hub import hf_hub_download
from sam2.build_sam import build_sam2
from sam2.automatic_mask_generator import SAM2AutomaticMaskGenerator

hf_hub_download(repo_id = "merve/sam2-hiera-base-plus", filename="sam2_hiera_base_plus.pt", local_dir = "./")
sam2_checkpoint = "../checkpoints/sam2_hiera_base_plus.pt"
model_cfg = "sam2_hiera_b+.yaml"

sam2 = build_sam2(model_cfg, sam2_checkpoint, device ='cuda', apply_postprocessing=False)

mask_generator = SAM2AutomaticMaskGenerator(sam2)
masks = mask_generator.generate(image)

📚 Documentation

The team behind SAM2 made example notebooks for all tasks.

See image predictor example for full example on prompting.
See automatic mask generation example for generating all masks.
See video object segmentation example

📄 License

This project is licensed under the Apache-2.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご