Image-Segmentation-DeepLabV3Plus

Passion Project

Image segmentation is the process of dividing an image into multiple meaningful and distinct regions or objects. It involves assigning a label or class to each pixel or region in an image to differentiate between different objects or areas of interest. This allows for the extraction and analysis of specific objects or regions within an image for various applications such as object recognition, scene understanding, and computer vision tasks. Through the use of TensorFlow, this project uses image segmentation to create masks that allow us to remove the background of an image.

Deep-Lab-V3 Plus Architecture

Training Parameters

1. batch_size = 2
2. learning_rate = 1e-4
3. epochs = 10

Training/Testing Results

Accuracy	F1	Jaccard	Recall	Precision
0.9205	0.8787	0.8092	0.9312	0.8645