segment anything(sam) into the pit

1. Related websites 

mirrors / facebookresearch / segment-anything · GitCode

This is a mirror, the download speed is fast

2. The pit used

1. Specify the original image to be segmented

 The red arrow above is pointing, it’s cheating, I thought your_image was a string.

no no

It should be as follows;

import cv2
import numpy as np
import matplotlib.pyplot as plt


from segment_anything import SamAutomaticMaskGenerator, sam_model_registry


def show_anns(anns):
    if len(anns) == 0:
        return
    sorted_anns = sorted(anns, key=(lambda x: x['area']), reverse=True)
    ax = plt.gca()
    ax.set_autoscale_on(False)
    polygons = []
    color = []
    for ann in sorted_anns:
        m = ann['segmentation']
        img = np.ones((m.shape[0], m.shape[1], 3))
        color_mask = np.random.random((1, 3)).tolist()[0]
        for i in range(3):
            img[:,:,i] = color_mask[i]
        ax.imshow(np.dstack((img, m*0.35)))


sam = sam_model_registry["default"](checkpoint="sam_vit_h_4b8939.pth")
device = "cuda"
sam.to(device=device)
mask_generator = SamAutomaticMaskGenerator(sam)

img = cv2.imread("1.jpg")
masks = mask_generator.generate(img)

plt.figure(figsize=(20,20))
plt.imshow(img)
show_anns(masks)
plt.axis('off')
plt.show()

The result is as follows:

 2, cv2 error

Cannot find reference ‘imread‘ in ‘__init__.py | __init__.py‘

 is the wrong version

I modified opencvthe version, instead 4.5.3.56, the problem is solved.

pip install -i https://pypi.douban.com/simple opencv-python==4.5.3.56 

3. The complete interactive segmentation code:

 If you want to interact and specify the foreground point, then the code is as follows:

import numpy as np
import torch
import matplotlib.pyplot as plt
import cv2

import sys
sys.path.append("..")
from segment_anything import sam_model_registry, SamPredictor


def show_mask(mask, ax, random_color=False):
    if random_color:
        color = np.concatenate([np.random.random(3), np.array([0.6])], axis=0)
    else:
        color = np.array([30 / 255, 144 / 255, 255 / 255, 0.6])
    h, w = mask.shape[-2:]
    mask_image = mask.reshape(h, w, 1) * color.reshape(1, 1, -1)
    ax.imshow(mask_image)


def show_points(coords, labels, ax, marker_size=375):
    pos_points = coords[labels == 1]
    neg_points = coords[labels == 0]
    ax.scatter(pos_points[:, 0], pos_points[:, 1], color='green', marker='*', s=marker_size, edgecolor='white',
               linewidth=1.25)
    ax.scatter(neg_points[:, 0], neg_points[:, 1], color='red', marker='*', s=marker_size, edgecolor='white',
               linewidth=1.25)


def show_box(box, ax):
    x0, y0 = box[0], box[1]
    w, h = box[2] - box[0], box[3] - box[1]
    ax.add_patch(plt.Rectangle((x0, y0), w, h, edgecolor='green', facecolor=(0, 0, 0, 0), lw=2))


image = cv2.imread('1.jpg')
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

plt.figure(figsize=(10,10))
plt.imshow(image)
plt.axis('on')
plt.show()


sam_checkpoint = "sam_vit_h_4b8939.pth"
model_type = "default"

device = "cuda"

sam = sam_model_registry[model_type](checkpoint=sam_checkpoint)
sam.to(device=device)

predictor = SamPredictor(sam)
predictor.set_image(image)


input_point = np.array([[494, 142]])
input_label = np.array([1])


plt.figure(figsize=(10,10))
plt.imshow(image)
show_points(input_point, input_label, plt.gca())
plt.axis('on')
plt.show()

masks, scores, logits = predictor.predict(
    point_coords=input_point,
    point_labels=input_label,
    multimask_output=True,
)

masks.shape  # (number_of_masks) x H x W

for i, (mask, score) in enumerate(zip(masks, scores)):
    plt.figure(figsize=(10,10))
    plt.imshow(image)
    show_mask(mask, plt.gca())
    show_points(input_point, input_label, plt.gca())
    plt.title(f"Mask {i+1}, Score: {score:.3f}", fontsize=18)
    plt.axis('off')
    plt.show()

The result is as follows: