Datasetpublic

HuggingFaceM4/Caltech-101

Pictures of objects belonging to 101 categories. About 40 to 800 images per category. Most categories have about 50 images. Collected in September 2003 by Fei-Fei Li, Marco Andreetto, and Marc'Aurelio Ranzato. The size of each image is roughly 300 x 200 pixels.

sourceHugging Facecc-by-4.0updated 2y agoView on Hugging Face

4likes623downloads

Dataset Card

Code snippet to visualise the position of the box

python

import matplotlib.image as img
import matplotlib.pyplot as plt
from datasets import load_dataset
from matplotlib.patches import Rectangle

# Load dataset
ds_name = "SaulLu/Caltech-101"
ds_config = "without_background_category"

ds_without = load_dataset(ds_name, ds_config, use_auth_token=True)

# Extract information for the sample we want to show
index = 100
sample = ds_without["train"][index]
box_coord = sample["annotation"]["box_coord"][0]
img_path = sample["image"].filename

# Create plot
# define Matplotlib figure and axis
fig, ax = plt.subplots()

# plot figure
image = img.imread(img_path)
ax.imshow(image)

# add rectangle to plot
ax.add_patch(
    Rectangle((box_coord[2], box_coord[0]), box_coord[3] - box_coord[2], box_coord[1] - box_coord[0], fill=None)
)

# display plot
plt.show()

Result: [image]