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
56
57
58
59
60
61
62
63
64
65
66
67
68
69

[package]
name = "object_detector"
description = "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. "
version = "0.5.0"
edition = "2024"
license = "AGPL-3.0"
homepage = "https://github.com/RuurdBijlsma/object_detector-rs"
repository = "https://github.com/RuurdBijlsma/object_detector-rs"
documentation = "https://docs.rs/object_detector"
exclude = ["assets", ".github", "TODO.md", "justfile", "output", "py-yolo"]

[dependencies]
ort = { version = "2.0.0-rc.12", features = ["ndarray"] }
ndarray = "0.17.2"
image = { version = "0.25.10", default-features = false }
serde_json = "1.0.149"
num_cpus = "1.17.0"
rayon = "1.11.0"
thiserror = "2.0.18"
bon = "3.9.1"
serde = { version = "1.0.228", optional = true, features = ["derive"] }
hf-hub = { version = "0.5.0", features = ["tokio"], optional = true }
open_clip_inference = { version = "0.4.0", optional = true }

[dev-dependencies]
criterion = { version = "0.8.2", features = ["html_reports"] }
serde = { version = "1.0", features = ["derive"] }
imageproc = "0.26.1"
ab_glyph = "0.2"
tokio = { version = "1.51", features = ["full"] }
color-eyre = "0.6.5"

[features]
default = ["download-binaries", "copy-dylibs", "hf-hub", "serde", "promptable"]
serde = ["dep:serde"]
hf-hub = ["dep:hf-hub"] # -> Enable downloading models from HuggingFace, relies on `tokio`
promptable = ["dep:open_clip_inference"]
# ort forwarded
download-binaries = ["ort/download-binaries", "open_clip_inference?/download-binaries"]
copy-dylibs = ["ort/copy-dylibs", "open_clip_inference?/copy-dylibs"]
load-dynamic = ["ort/load-dynamic", "open_clip_inference?/load-dynamic"]
# ort execution providers
cuda = ["ort/cuda", "open_clip_inference?/cuda"]
tensorrt = ["ort/tensorrt", "open_clip_inference?/tensorrt"]
nvrtx = ["ort/nvrtx", "open_clip_inference?/nvrtx"]
xnnpack = ["ort/xnnpack", "open_clip_inference?/xnnpack"]
webgpu = ["ort/webgpu", "open_clip_inference?/webgpu"]
directml = ["ort/directml", "open_clip_inference?/directml"]
coreml = ["ort/coreml", "open_clip_inference?/coreml"]
migraphx = ["ort/migraphx", "open_clip_inference?/migraphx"]
openvino = ["ort/openvino", "open_clip_inference?/openvino"]
onednn = ["ort/onednn", "open_clip_inference?/onednn"]
qnn = ["ort/qnn", "open_clip_inference?/qnn"]
cann = ["ort/cann", "open_clip_inference?/cann"]
nnapi = ["ort/nnapi", "open_clip_inference?/nnapi"]
tvm = ["ort/tvm", "open_clip_inference?/tvm"]
acl = ["ort/acl", "open_clip_inference?/acl"]
armnn = ["ort/armnn", "open_clip_inference?/armnn"]
vitis = ["ort/vitis", "open_clip_inference?/vitis"]
rknpu = ["ort/rknpu", "open_clip_inference?/rknpu"]
azure = ["ort/azure", "open_clip_inference?/azure"]

[[bench]]
name = "component_benchmarks"
harness = false

[[bench]]
name = "full_benchmarks"
harness = false