yscv-cli 0.1.7

CLI for inference, benchmarking, and evaluation
Documentation 
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
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160

use yscv_detect::BoundingBox;
use yscv_recognize::Recognizer;
use yscv_tensor::Tensor;
use yscv_video::{
    CameraConfig, CameraFrameSource, Frame, FrameSource, InMemoryFrameSource, resolve_camera_device,
};

use crate::config::{CliConfig, DetectTarget};
use crate::error::AppError;
use crate::util::embedding_from_bbox;

pub fn open_camera_source(cli: &CliConfig) -> Result<CameraFrameSource, AppError> {
    let device_index = if let Some(query) = cli.device_name_query.as_deref() {
        let selected = resolve_camera_device(query)?;
        println!(
            "camera device resolved: query=`{}` -> {}: {}",
            query, selected.index, selected.label
        );
        selected.index
    } else {
        cli.device_index
    };

    let source = CameraFrameSource::open(CameraConfig {
        device_index,
        width: cli.width,
        height: cli.height,
        fps: cli.fps,
    })?;
    Ok(source)
}

pub fn build_source(
    cli: &CliConfig,
    recognizer: &mut Recognizer,
) -> Result<Box<dyn FrameSource>, AppError> {
    if cli.camera {
        let source = open_camera_source(cli)?;
        return Ok(Box::new(source));
    }

    let frame_height = 6usize;
    let frame_width = 6usize;
    let frames = match cli.detect_target {
        DetectTarget::People => vec![
            frame_with_boxes(0, 0, frame_height, frame_width, &[(1, 1, 3, 3, 0.9)])?,
            frame_with_boxes(1, 33_333, frame_height, frame_width, &[(1, 2, 3, 4, 0.85)])?,
            frame_with_boxes(
                2,
                66_666,
                frame_height,
                frame_width,
                &[(1, 3, 3, 5, 0.9), (4, 0, 6, 2, 0.88)],
            )?,
        ],
        DetectTarget::Faces => vec![
            frame_with_rgb_boxes(
                0,
                0,
                frame_height,
                frame_width,
                &[(1, 1, 3, 3, [0.78, 0.60, 0.46])],
                [0.10, 0.10, 0.12],
            )?,
            frame_with_rgb_boxes(
                1,
                33_333,
                frame_height,
                frame_width,
                &[(1, 2, 3, 4, [0.79, 0.61, 0.47])],
                [0.10, 0.10, 0.12],
            )?,
            frame_with_rgb_boxes(
                2,
                66_666,
                frame_height,
                frame_width,
                &[
                    (1, 3, 3, 5, [0.78, 0.60, 0.46]),
                    (4, 0, 6, 2, [0.72, 0.55, 0.41]),
                ],
                [0.10, 0.10, 0.12],
            )?,
        ],
    };

    if recognizer.identities().is_empty() {
        let known_embedding = embedding_from_bbox(
            BoundingBox {
                x1: 1.0,
                y1: 1.0,
                x2: 3.0,
                y2: 3.0,
            },
            frame_width as f32,
            frame_height as f32,
        )?;
        recognizer.enroll("alice", known_embedding)?;
    }

    Ok(Box::new(InMemoryFrameSource::new(frames)))
}

fn frame_with_boxes(
    index: u64,
    ts_us: u64,
    height: usize,
    width: usize,
    boxes: &[(usize, usize, usize, usize, f32)],
) -> Result<Frame, AppError> {
    let mut data = vec![0.0f32; height * width];
    for (x1, y1, x2, y2, value) in boxes {
        for y in *y1..*y2 {
            for x in *x1..*x2 {
                data[y * width + x] = *value;
            }
        }
    }
    Ok(Frame::new(
        index,
        ts_us,
        Tensor::from_vec(vec![height, width, 1], data)?,
    )?)
}

fn frame_with_rgb_boxes(
    index: u64,
    ts_us: u64,
    height: usize,
    width: usize,
    boxes: &[(usize, usize, usize, usize, [f32; 3])],
    background_rgb: [f32; 3],
) -> Result<Frame, AppError> {
    let mut data = vec![0.0f32; height * width * 3];
    for y in 0..height {
        for x in 0..width {
            let base = (y * width + x) * 3;
            data[base] = background_rgb[0];
            data[base + 1] = background_rgb[1];
            data[base + 2] = background_rgb[2];
        }
    }

    for (x1, y1, x2, y2, rgb) in boxes {
        for y in *y1..*y2 {
            for x in *x1..*x2 {
                let base = (y * width + x) * 3;
                data[base] = rgb[0];
                data[base + 1] = rgb[1];
                data[base + 2] = rgb[2];
            }
        }
    }

    Ok(Frame::new(
        index,
        ts_us,
        Tensor::from_vec(vec![height, width, 3], data)?,
    )?)
}