object_detector 0.5.0

Object detection using ORT and the yoloe-26-seg model. This model can detect multiple objects per image, each having a tag, pixel-level mask, and a boundingbox. It's pretrained, it has a vocabulary of 4000+ 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

use thiserror::Error;

#[cfg(feature = "hf-hub")]
use hf_hub::api::tokio::ApiError;

#[derive(Error, Debug)]
pub enum ObjectDetectorError {
    #[error("IO Error: {0}")]
    Io(#[from] std::io::Error),

    #[error("Image Error: {0}")]
    Image(#[from] image::ImageError),

    #[error("JSON Error: {0}")]
    Json(#[from] serde_json::Error),

    #[error("ONNX Runtime Error: {0}")]
    Ort(String),

    #[error("NdArray Error: {0}")]
    NdArray(#[from] ndarray::ShapeError),

    #[cfg(feature = "promptable")]
    #[error("Clip Error: {0}")]
    Clip(#[from] open_clip_inference::ClipError),

    #[cfg(feature = "hf-hub")]
    #[error("Hugging Face Hub error: {0}")]
    HfHub(String),

    #[error("Model Consistency Error: {0}")]
    InvalidModel(String),

    #[error("Lock poison error: {0}")]
    LockPoison(String),
}

impl<'a, T> From<std::sync::PoisonError<std::sync::MutexGuard<'a, T>>> for ObjectDetectorError {
    fn from(err: std::sync::PoisonError<std::sync::MutexGuard<'a, T>>) -> Self {
        Self::LockPoison(err.to_string())
    }
}

impl<T> From<ort::Error<T>> for ObjectDetectorError {
    fn from(err: ort::Error<T>) -> Self {
        Self::Ort(err.to_string())
    }
}

#[cfg(feature = "hf-hub")]
impl From<ApiError> for ObjectDetectorError {
    fn from(value: ApiError) -> Self {
        Self::HfHub(value.to_string())
    }
}