Kalosm Vision

Kalosm Vision是Kalosm框架的一组图像模型和实用工具。它包括从文本生成图像和将图像分割成对象的工具。

图像生成

您可以使用Wuerstchen模型从文本生成图像

use futures_util::StreamExt;
use kalosm_vision::{Wuerstchen, WuerstchenInferenceSettings};

#[tokio::main]
async fn main() {
    let model = Wuerstchen::builder().build().await.unwrap();
    let settings = WuerstchenInferenceSettings::new(
        "a cute cat with a hat in a room covered with fur with incredible detail",
    );

    if let Ok(mut images) = model.run(settings) {
        while let Some(image) = images.next().await {
            if let Some(buf) = image.generated_image() {
                buf.save(&format!("{}.png", image.sample_num())).unwrap();
            }
        }
    }
}

图像分割

Kalosm支持使用SegmentAnything模型进行图像分割。您可以使用SegmentAnything::segment_everything方法将图像分割成对象，或使用SegmentAnything::segment_from_points方法在特定点将图像分割成对象

use kalosm::vision::*;

let model = SegmentAnything::builder().build().unwrap();
let image = image::open("examples/landscape.jpg").unwrap();
let x = image.width() / 2;
let y = image.height() / 4;
let images = model
    .segment_from_points(
        SegmentAnythingInferenceSettings::new(image)
            .unwrap()
            .add_goal_point(x, y),
    )
    .unwrap();

images.save("out.png").unwrap();