Datasets:

ArXiv:
Tags:
art
License:
File size: 1,803 Bytes
ce9cc74
 
 
9c206a9
ce9cc74
 
9c206a9
9a77a6e
e841ff0
9a77a6e
f8381ae
e841ff0
9c206a9
 
 
41831b9
 
9c206a9
 
 
e841ff0
9c206a9
e841ff0
9c206a9
 
 
 
 
 
 
 
 
e841ff0
af6cbf1
 
 
 
9c206a9
af6cbf1
9c206a9
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
---
license: apache-2.0
tags:
  - art
pretty_name: Human Segmentation Dataset
---

# Human Segmentation Dataset

This dataset was created **for developing the best fully open-source background remover** of images with humans.
The dataset was crafted with [LayerDiffuse](https://github.com/layerdiffusion/LayerDiffuse), a Stable Diffusion extension for generating transparent images.

The dataset covers a diverse set of humans: various skin tones, clothes, hair styles etc.
Since Stable Diffusion is not perfect, the dataset contains images with flaws. Still the dataset is good enough for training background remover models.

The resulting model will be similar to [RMBG-1.4](https://huggingface.co/briaai/RMBG-1.4), but with open training data/process and commercially free to use.

I had some trouble with the Hugging Face file upload. You can find the data here: [GDrvie](https://drive.google.com/drive/folders/1K1lK6nSoaQ7PLta-bcfol3XSGZA1b9nt?usp=drive_link)

The dataset contains transparent images of humans (`/humans`) which are randomly combined with backgrounds (`/backgrounds`). Then the ground truth (`/gt`) for segmentation was computed based on the transparent images. The results are written to a training and validation dataset.

I created more than 5.000 images with people and more than 5.000 diverse backgrounds.

# Create Training Dataset

The following scripts created training and validation data. Adding to this data is augmented.

Notice: download the dataset from [GDrvie](https://drive.google.com/drive/folders/1K1lK6nSoaQ7PLta-bcfol3XSGZA1b9nt?usp=drive_link).

```
./create_dataset.sh
```

# Support

If you identify weaknesses in the data, please contact me.

# Changelog

- Added more diverse backgrounds (natural landscapes, streets, houses)
- Added more close-up images