cumulus3d-core 0.11.0

Core types for cumulus3d: camera models, rigid/similarity transforms, reconstruction graph, feature store, RANSAC, PLY/GPS I/O
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

/*
 * analyzer.rs
 *
 * Copyright (c) 2026 ParkSangWoo
 *
 * Author(s):
 *
 *      ParkSangWoo <dev.parksangwoo@gmail.com>
 *
 *
 * This file is part of cumulus3d.
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 *
 * Third-party notices: parts of the algorithms, default parameters and data
 * formats in this file follow other open-source projects. Their copyright
 * notices and licenses are reproduced in THIRD_PARTY_NOTICES.md.
 *
 * SPDX-License-Identifier: Apache-2.0
 */

//! Model statistics (output of the model_analyzer subcommand).

use crate::io::fmt::format_g;
use crate::reconstruction::Reconstruction;

/// All items of model_analyzer.
#[derive(Clone, Debug, PartialEq)]
pub struct ModelStats {
    /// Number of rigs.
    pub num_rigs: usize,
    /// Number of cameras.
    pub num_cameras: usize,
    /// Number of frames.
    pub num_frames: usize,
    /// Number of registered frames.
    pub registered_frame_count: usize,
    /// Number of images.
    pub num_images: usize,
    /// Number of registered images.
    pub registered_image_count: usize,
    /// Number of 3D points.
    pub num_points3d: usize,
    /// Sum over registered images of "number of 2D points linked to a 3D point".
    pub num_observations: usize,
    /// Mean track length.
    pub mean_track_length: f64,
    /// Mean number of observations per registered image.
    pub mean_observations_per_image: f64,
    /// Mean of the stored point errors (excluding -1). Not recomputed.
    pub mean_reprojection_error: f64,
}

impl ModelStats {
    /// Computes the statistics from a reconstruction.
    pub fn compute(rec: &Reconstruction) -> Self {
        Self {
            num_rigs: rec.num_rigs(),
            num_cameras: rec.num_cameras(),
            num_frames: rec.num_frames(),
            registered_frame_count: rec.registered_frame_count(),
            num_images: rec.num_images(),
            registered_image_count: rec.registered_image_count(),
            num_points3d: rec.num_points3d(),
            num_observations: rec.total_observations(),
            mean_track_length: rec.mean_track_len(),
            mean_observations_per_image: rec.mean_obs_per_registered_image(),
            mean_reprojection_error: rec.mean_reproj_error(),
        }
    }

    /// Fixed item order and wording (log prefix excluded). Integers use %d, reals use %f.
    pub fn lines(&self) -> Vec<String> {
        vec![
            format!("Rigs: {}", self.num_rigs),
            format!("Cameras: {}", self.num_cameras),
            format!("Frames: {}", self.num_frames),
            format!("Registered frames: {}", self.registered_frame_count),
            format!("Images: {}", self.num_images),
            format!("Registered images: {}", self.registered_image_count),
            format!("Points: {}", self.num_points3d),
            format!("Observations: {}", self.num_observations),
            format!("Mean track length: {:.6}", self.mean_track_length),
            format!("Mean observations per image: {:.6}", self.mean_observations_per_image),
            format!("Mean reprojection error: {:.6}px", self.mean_reprojection_error),
        ]
    }
}

/// model_analyzer output lines. If verbose, the camera and registered-image lists are appended.
pub fn analyzer_lines(rec: &Reconstruction, verbose: bool) -> Vec<String> {
    let mut out = ModelStats::compute(rec).lines();
    if verbose {
        out.push("--- Cameras ---".into());
        for c in rec.cameras().values() {
            // Design decision: parameter numbers are formatted with 6 significant digits (%g).
            let params = c.params.iter().map(|p| format_g(*p, 6)).collect::<Vec<_>>().join(", ");
            out.push(format!(" - Camera Id: {}, Model Name: {}, Params: {}", c.camera_id, c.model.name(), params));
        }
        out.push("--- Images ---".into());
        for id in rec.registered_images() {
            let name = rec.image(id).map(|im| im.name.as_str()).unwrap_or("");
            out.push(format!(" - Registered Image Id: {id}, Name: {name}"));
        }
    }
    out
}

#[cfg(test)]
mod tests {
    use super::*;
    use crate::camera::{Camera, CameraModelKind};
    use crate::geometry::{Rigid3, Vec2, Vec3};
    use crate::reconstruction::{Image, TrackEntry};

    #[test]
    fn small_model_stats() {
        let mut r = Reconstruction::new();
        let mut cam = Camera::from_focal(CameraModelKind::OpenCv, 100.0, 200, 100);
        cam.camera_id = 1;
        r.add_camera_own_rig(cam).unwrap();
        for i in 1..=3u32 {
            let pts = (0..4).map(|k| Vec2::new(k as f64, 0.0));
            r.add_image_own_frame(Image::new(i, format!("i{i}.jpg"), 1, pts), Some(Rigid3::identity())).unwrap();
            r.register_image(i).unwrap();
        }
        let t = |i, k| TrackEntry::new(i, k);
        let tracks = [vec![t(1, 0), t(2, 0)], vec![t(1, 1), t(2, 1), t(3, 1)], vec![t(2, 2), t(3, 2)], vec![t(1, 3), t(3, 3)]];
        for (tr, e) in tracks.into_iter().zip([1.0, 2.0, -1.0, 3.0]) {
            let id = r.add_point3d(Vec3::new(0.0, 0.0, 1.0), tr, [0; 3]).unwrap();
            r.set_point3d_error(id, e).unwrap();
        }
        let s = ModelStats::compute(&r);
        assert_eq!(s.num_observations, 9);
        assert_eq!(s.mean_track_length, 2.25);
        assert_eq!(s.mean_observations_per_image, 3.0);
        assert_eq!(s.mean_reprojection_error, 2.0);
        let l = analyzer_lines(&r, true);
        assert_eq!(
            &l[..11],
            &[
                "Rigs: 1",
                "Cameras: 1",
                "Frames: 3",
                "Registered frames: 3",
                "Images: 3",
                "Registered images: 3",
                "Points: 4",
                "Observations: 9",
                "Mean track length: 2.250000",
                "Mean observations per image: 3.000000",
                "Mean reprojection error: 2.000000px",
            ]
        );
        assert_eq!(l[11], "--- Cameras ---");
        assert_eq!(l[12], " - Camera Id: 1, Model Name: OPENCV, Params: 100, 100, 100, 50, 0, 0, 0, 0");
        assert_eq!(l[14], " - Registered Image Id: 1, Name: i1.jpg");
        let empty = ModelStats::compute(&Reconstruction::new());
        assert_eq!(empty.lines()[10], "Mean reprojection error: 0.000000px");
    }
}