cameras 0.3.1

A cross-platform camera library for Rust, built with data-oriented design. Explicit format negotiation, push-based frame delivery, typed errors, and zero trait objects.
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
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285

//! Pixel format conversion helpers.
//!
//! Frames come out of the library in whatever native format the platform delivered.
//! These functions convert them to the two formats most consumers want: packed RGB8
//! (24 bits per pixel) and packed RGBA8 (32 bits per pixel). Stride is honored, so the
//! output is always tightly packed regardless of how padded the source buffer was.

use crate::error::Error;
use crate::types::{Frame, PixelFormat};

/// Convert a frame to packed 24-bit RGB (3 bytes per pixel).
///
/// Supports every [`PixelFormat`]. MJPEG is decoded via `zune-jpeg`.
pub fn to_rgb8(frame: &Frame) -> Result<Vec<u8>, Error> {
    let width = frame.width as usize;
    let height = frame.height as usize;
    let stride = frame.stride as usize;
    match frame.pixel_format {
        PixelFormat::Rgb8 => Ok(frame.plane_primary.to_vec()),
        PixelFormat::Rgba8 => Ok(rgba_to_rgb(&frame.plane_primary)),
        PixelFormat::Bgra8 => Ok(bgra_to_rgb(&frame.plane_primary, width, height, stride)),
        PixelFormat::Yuyv => Ok(yuyv_to_rgb(&frame.plane_primary, width, height, stride)),
        PixelFormat::Nv12 => Ok(nv12_to_rgb(
            &frame.plane_primary,
            &frame.plane_secondary,
            width,
            height,
            stride,
        )),
        PixelFormat::Mjpeg => mjpeg_to_rgb(&frame.plane_primary),
    }
}

/// Convert a frame to packed 32-bit RGBA (4 bytes per pixel).
///
/// For formats without an alpha channel (RGB8, YUYV, NV12, MJPEG), the alpha byte is
/// filled with 0xFF. For BGRA8, the channel order is swapped in place.
pub fn to_rgba8(frame: &Frame) -> Result<Vec<u8>, Error> {
    let width = frame.width as usize;
    let height = frame.height as usize;
    let stride = frame.stride as usize;
    match frame.pixel_format {
        PixelFormat::Rgba8 => Ok(frame.plane_primary.to_vec()),
        PixelFormat::Bgra8 => Ok(bgra_to_rgba(&frame.plane_primary, width, height, stride)),
        _ => {
            let rgb = to_rgb8(frame)?;
            Ok(rgb_to_rgba(&rgb))
        }
    }
}

fn rgba_to_rgb(data: &[u8]) -> Vec<u8> {
    let mut output = Vec::with_capacity((data.len() / 4) * 3);
    for chunk in data.chunks_exact(4) {
        output.push(chunk[0]);
        output.push(chunk[1]);
        output.push(chunk[2]);
    }
    output
}

fn bgra_to_rgb(data: &[u8], width: usize, height: usize, stride: usize) -> Vec<u8> {
    let effective_stride = if stride == 0 { width * 4 } else { stride };
    let rows_available = data.len().checked_div(effective_stride).unwrap_or(0);
    let rows = height.min(rows_available);
    let row_bytes_wanted = width * 4;
    let mut output = Vec::with_capacity(rows * width * 3);
    for row in 0..rows {
        let offset = row * effective_stride;
        let end = offset.saturating_add(row_bytes_wanted).min(data.len());
        let row_bytes = &data[offset..end];
        for pixel in row_bytes.chunks_exact(4) {
            output.push(pixel[2]);
            output.push(pixel[1]);
            output.push(pixel[0]);
        }
    }
    output
}

fn bgra_to_rgba(data: &[u8], width: usize, height: usize, stride: usize) -> Vec<u8> {
    let effective_stride = if stride == 0 { width * 4 } else { stride };
    let rows_available = data.len().checked_div(effective_stride).unwrap_or(0);
    let rows = height.min(rows_available);
    let row_bytes_wanted = width * 4;
    let mut output = Vec::with_capacity(width * height * 4);
    for row in 0..rows {
        let offset = row * effective_stride;
        let end = offset.saturating_add(row_bytes_wanted).min(data.len());
        let row_bytes = &data[offset..end];
        for chunk in row_bytes.chunks_exact(4) {
            output.push(chunk[2]);
            output.push(chunk[1]);
            output.push(chunk[0]);
            // Force opaque. Windows Media Foundation delivers BGRA with
            // alpha byte set to 0 (XRGB semantics), which makes the frame
            // render fully transparent in consumers that respect alpha.
            output.push(0xFF);
        }
    }
    output
}

fn rgb_to_rgba(data: &[u8]) -> Vec<u8> {
    let mut output = Vec::with_capacity((data.len() / 3) * 4);
    for chunk in data.chunks_exact(3) {
        output.push(chunk[0]);
        output.push(chunk[1]);
        output.push(chunk[2]);
        output.push(255);
    }
    output
}

fn yuyv_to_rgb(data: &[u8], width: usize, height: usize, stride: usize) -> Vec<u8> {
    let row_bytes = width * 2;
    let effective_stride = if stride == 0 { row_bytes } else { stride };
    let mut output = Vec::with_capacity(width * height * 3);
    for row in 0..height {
        let start = row * effective_stride;
        let row_slice = &data[start..start + row_bytes];
        for chunk in row_slice.chunks_exact(4) {
            let y0 = chunk[0] as i32;
            let u = chunk[1] as i32 - 128;
            let y1 = chunk[2] as i32;
            let v = chunk[3] as i32 - 128;
            push_yuv_rgb(&mut output, y0, u, v);
            push_yuv_rgb(&mut output, y1, u, v);
        }
    }
    output
}

fn nv12_to_rgb(
    y_plane: &[u8],
    uv_plane: &[u8],
    width: usize,
    height: usize,
    stride: usize,
) -> Vec<u8> {
    let y_stride = if stride == 0 { width } else { stride };
    let uv_stride = y_stride;
    let mut output = vec![0u8; width * height * 3];
    for row in 0..height {
        for col in 0..width {
            let y = y_plane[row * y_stride + col] as i32;
            let uv_row = row / 2;
            let uv_col = (col / 2) * 2;
            let uv_index = uv_row * uv_stride + uv_col;
            let u = uv_plane[uv_index] as i32 - 128;
            let v = uv_plane[uv_index + 1] as i32 - 128;
            let base = (row * width + col) * 3;
            let (r, g, b) = yuv_to_rgb(y, u, v);
            output[base] = r;
            output[base + 1] = g;
            output[base + 2] = b;
        }
    }
    output
}

fn push_yuv_rgb(output: &mut Vec<u8>, y: i32, u: i32, v: i32) {
    let (r, g, b) = yuv_to_rgb(y, u, v);
    output.push(r);
    output.push(g);
    output.push(b);
}

fn yuv_to_rgb(y: i32, u: i32, v: i32) -> (u8, u8, u8) {
    let c = y - 16;
    let d = u;
    let e = v;
    let r = ((298 * c + 409 * e + 128) >> 8).clamp(0, 255) as u8;
    let g = ((298 * c - 100 * d - 208 * e + 128) >> 8).clamp(0, 255) as u8;
    let b = ((298 * c + 516 * d + 128) >> 8).clamp(0, 255) as u8;
    (r, g, b)
}

fn mjpeg_to_rgb(data: &[u8]) -> Result<Vec<u8>, Error> {
    let mut decoder = zune_jpeg::JpegDecoder::new(data);
    decoder
        .decode()
        .map_err(|error| Error::MjpegDecode(error.to_string()))
}

#[cfg(feature = "analysis")]
pub(crate) fn to_luma8(frame: &Frame) -> Vec<u8> {
    let width = frame.width as usize;
    let height = frame.height as usize;
    let stride = frame.stride as usize;
    match frame.pixel_format {
        PixelFormat::Rgb8 => rgb_to_luma(&frame.plane_primary),
        PixelFormat::Rgba8 => rgba_to_luma(&frame.plane_primary),
        PixelFormat::Bgra8 => bgra_to_luma(&frame.plane_primary, width, height, stride),
        PixelFormat::Yuyv => yuyv_to_luma(&frame.plane_primary, width, height, stride),
        PixelFormat::Nv12 => nv12_y_to_luma(&frame.plane_primary, width, height, stride),
        PixelFormat::Mjpeg => mjpeg_to_rgb(&frame.plane_primary)
            .map(|rgb| rgb_to_luma(&rgb))
            .unwrap_or_default(),
    }
}

#[cfg(feature = "analysis")]
const LUMA_WEIGHT_RED: u32 = 299;
#[cfg(feature = "analysis")]
const LUMA_WEIGHT_GREEN: u32 = 587;
#[cfg(feature = "analysis")]
const LUMA_WEIGHT_BLUE: u32 = 114;

#[cfg(feature = "analysis")]
fn rec601_luma(red: u8, green: u8, blue: u8) -> u8 {
    let weighted = LUMA_WEIGHT_RED * red as u32
        + LUMA_WEIGHT_GREEN * green as u32
        + LUMA_WEIGHT_BLUE * blue as u32;
    (weighted / 1000) as u8
}

#[cfg(feature = "analysis")]
fn rgb_to_luma(data: &[u8]) -> Vec<u8> {
    let mut output = Vec::with_capacity(data.len() / 3);
    for pixel in data.chunks_exact(3) {
        output.push(rec601_luma(pixel[0], pixel[1], pixel[2]));
    }
    output
}

#[cfg(feature = "analysis")]
fn rgba_to_luma(data: &[u8]) -> Vec<u8> {
    let mut output = Vec::with_capacity(data.len() / 4);
    for pixel in data.chunks_exact(4) {
        output.push(rec601_luma(pixel[0], pixel[1], pixel[2]));
    }
    output
}

#[cfg(feature = "analysis")]
fn bgra_to_luma(data: &[u8], width: usize, height: usize, stride: usize) -> Vec<u8> {
    let effective_stride = if stride == 0 { width * 4 } else { stride };
    let rows_available = data.len().checked_div(effective_stride).unwrap_or(0);
    let rows = height.min(rows_available);
    let row_bytes_wanted = width * 4;
    let mut output = Vec::with_capacity(rows * width);
    for row in 0..rows {
        let offset = row * effective_stride;
        let end = offset.saturating_add(row_bytes_wanted).min(data.len());
        let row_bytes = &data[offset..end];
        for pixel in row_bytes.chunks_exact(4) {
            output.push(rec601_luma(pixel[2], pixel[1], pixel[0]));
        }
    }
    output
}

#[cfg(feature = "analysis")]
fn yuyv_to_luma(data: &[u8], width: usize, height: usize, stride: usize) -> Vec<u8> {
    let row_bytes = width * 2;
    let effective_stride = if stride == 0 { row_bytes } else { stride };
    let rows_available = data.len().checked_div(effective_stride).unwrap_or(0);
    let rows = height.min(rows_available);
    let mut output = Vec::with_capacity(rows * width);
    for row in 0..rows {
        let start = row * effective_stride;
        let end = start.saturating_add(row_bytes).min(data.len());
        let row_slice = &data[start..end];
        for pair in row_slice.chunks_exact(4) {
            output.push(pair[0]);
            output.push(pair[2]);
        }
    }
    output
}

#[cfg(feature = "analysis")]
fn nv12_y_to_luma(y_plane: &[u8], width: usize, height: usize, stride: usize) -> Vec<u8> {
    let y_stride = if stride == 0 { width } else { stride };
    let rows_available = y_plane.len().checked_div(y_stride).unwrap_or(0);
    let rows = height.min(rows_available);
    let mut output = Vec::with_capacity(rows * width);
    for row in 0..rows {
        let start = row * y_stride;
        let end = start.saturating_add(width).min(y_plane.len());
        output.extend_from_slice(&y_plane[start..end]);
    }
    output
}