Compare commits

...
34 Commits
Author SHA1 Message Date
SkalskiP 19396771f1 bump package version from 0.21.0rc4 to 0.21.0rc5 2024-05-13 15:58:22 +02:00
Onuralp SEZER f5324897a3 Merge pull request #1189 from roboflow/dependabot/pip/mkdocs-material-9.5.22
⬆️ Bump mkdocs-material from 9.5.21 to 9.5.22
2024-05-13 16:56:57 +03:00
Piotr Skalski cd8a2be04b Merge pull request #1178 from roboflow/feat/inference-slicer-segmentation
Feat/inference slicer segmentation
2024-05-13 15:55:02 +02:00
Linas Kondrackis 8901192b92 ruff 2024-05-13 16:53:06 +03:00
Linas Kondrackis a9044e4fed Merge branch 'develop' into feat/inference-slicer-segmentation 2024-05-13 16:52:50 +03:00
Linas Kondrackis 85e4c0ec65 Slicer: highlight & mask docstring 2024-05-13 16:51:59 +03:00
Linas Kondrackis 93398108ee Slier docs: remove transfomer example - has issues. 2024-05-13 16:26:35 +03:00
Piotr Skalski f41adca384 Merge pull request #1173 from roboflow/vertex_label_annotator
`VertexLabelAnnotator`
2024-05-13 15:23:51 +02:00
SkalskiP d64bb3879c Merge remote-tracking branch 'origin/vertex_label_annotator' into vertex_label_annotator 2024-05-13 15:21:35 +02:00
SkalskiP 452e1f1ea4 more VertexLabelAnnotator docs 2024-05-13 15:21:25 +02:00
Linas Kondrackis 45e5f49c5a Infrence slicer docs: fix callbacks in transfomers 2024-05-13 16:20:05 +03:00
LinasKo 4c36b39f94 Untested: Small object segmentationation 2024-05-13 15:41:43 +03:00
pre-commit-ci[bot] e13c7dc22c fix(pre_commit): 🎨 auto format pre-commit hooks 2024-05-13 12:31:37 +00:00
SkalskiP 713719a582 Merge remote-tracking branch 'origin/vertex_label_annotator' into vertex_label_annotator 2024-05-13 14:31:12 +02:00
SkalskiP 051fe5178b VertexLabelAnnotator plugged into docs 2024-05-13 14:30:55 +02:00
pre-commit-ci[bot] 1ebccebf4c fix(pre_commit): 🎨 auto format pre-commit hooks 2024-05-13 12:01:26 +00:00
SkalskiP f56d7a1733 changes after code review 2024-05-13 14:01:07 +02:00
LinasKo 82ab6c57cb infrenceSlicer: resolution_wh 2024-05-13 14:14:42 +03:00
Linas Kondrackis 816aecb51b PR comments: tuples, docs, optional 2024-05-13 11:35:23 +03:00
dependabot[bot] 3c8995758d ⬆️ Bump mkdocs-material from 9.5.21 to 9.5.22
Bumps [mkdocs-material](https://github.com/squidfunk/mkdocs-material) from 9.5.21 to 9.5.22.
- [Release notes](https://github.com/squidfunk/mkdocs-material/releases)
- [Changelog](https://github.com/squidfunk/mkdocs-material/blob/master/CHANGELOG)
- [Commits](https://github.com/squidfunk/mkdocs-material/compare/9.5.21...9.5.22)

---
updated-dependencies:
- dependency-name: mkdocs-material
  dependency-type: direct:development
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
2024-05-13 01:21:27 +00:00
Onuralp SEZER acfabc247f Merge pull request #1185 from roboflow/dependabot/pip/ruff-0.4.4
⬆️ Bump ruff from 0.4.3 to 0.4.4
2024-05-11 23:54:29 +03:00
Onuralp SEZER 61a63c83d9 Merge pull request #1188 from roboflow/docs/missing-return-in-slicer-callback
Minor docs update: missing return in inference slicer callback
2024-05-11 21:56:13 +03:00
Linas Kondrackis 528c960587 Minor docs update: missing return in inference slicer callback 2024-05-11 13:38:59 +03:00
Linas Kondrackis 6c6171b5d5 Simpify passing of shape 2024-05-10 18:25:02 +03:00
Linas Kondrackis b8efa25e93 Merge branch 'develop' into feat/inference-slicer-segmentation 2024-05-10 17:18:07 +03:00
dependabot[bot] 3b61ee674e ⬆️ Bump ruff from 0.4.3 to 0.4.4
Bumps [ruff](https://github.com/astral-sh/ruff) from 0.4.3 to 0.4.4.
- [Release notes](https://github.com/astral-sh/ruff/releases)
- [Changelog](https://github.com/astral-sh/ruff/blob/main/CHANGELOG.md)
- [Commits](https://github.com/astral-sh/ruff/compare/v0.4.3...v0.4.4)

---
updated-dependencies:
- dependency-name: ruff
  dependency-type: direct:development
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
2024-05-10 00:28:34 +00:00
Linas Kondrackis 45563dfddf InferenceSlicer: Now works with segmentation 2024-05-08 14:07:37 +03:00
pre-commit-ci[bot] 7823e0de89 fix(pre_commit): 🎨 auto format pre-commit hooks 2024-05-06 16:16:39 +00:00
SkalskiP ab0882dd55 initial version of VertexLabelAnnotator 2024-05-06 18:15:05 +02:00
SkalskiP ec6864ff04 roll back new draw_rounded_rectangle implementation 2024-05-06 18:00:13 +02:00
SkalskiP 81786ff9d3 Merge remote-tracking branch 'origin/vertex_label_annotator' into vertex_label_annotator 2024-05-06 17:40:43 +02:00
SkalskiP 95c243c058 better draw_rounded_rectangle implementation 2024-05-06 17:40:33 +02:00
pre-commit-ci[bot] 5c539c3bda fix(pre_commit): 🎨 auto format pre-commit hooks 2024-05-06 13:49:24 +00:00
SkalskiP 5d38120743 added from_xyxy and as_xyxy_int_tuple methods to Rect. added draw_rounded_rectangle function. 2024-05-06 15:39:59 +02:00
11 changed files with 554 additions and 44 deletions
+18
View File
@@ -65,8 +65,26 @@ status: new
:::supervision.detection.utils.move_boxes
<div class="md-typeset">
<h2><a href="#supervision.detection.utils.move_masks">move_masks</a></h2>
</div>
:::supervision.detection.utils.move_masks
<div class="md-typeset">
<h2><a href="#supervision.detection.utils.scale_boxes">scale_boxes</a></h2>
</div>
:::supervision.detection.utils.scale_boxes
<div class="md-typeset">
<h2><a href="#supervision.detection.utils.clip_boxes">clip_boxes</a></h2>
</div>
:::supervision.detection.utils.clip_boxes
<div class="md-typeset">
<h2><a href="#supervision.detection.utils.pad_boxes">pad_boxes</a></h2>
</div>
:::supervision.detection.utils.pad_boxes
+68 -8
View File
@@ -6,7 +6,7 @@ status: new
# Detect Small Objects
This guide shows how to detect small objects
with the [Inference](https://github.com/roboflow/inference),
with the [Inference](https://github.com/roboflow/inference),
[Ultralytics](https://github.com/ultralytics/ultralytics) or
[Transformers](https://github.com/huggingface/transformers) packages using
[`InferenceSlicer`](/latest/detection/tools/inference_slicer/#supervision.detection.tools.inference_slicer.InferenceSlicer).
@@ -68,10 +68,10 @@ size relative to the image resolution.
import torch
import supervision as sv
from PIL import Image
from transformers import DetrImageProcessor, DetrForObjectDetection
from transformers import DetrImageProcessor, DetrForSegmentation
processor = DetrImageProcessor.from_pretrained("facebook/detr-resnet-50")
model = DetrForObjectDetection.from_pretrained("facebook/detr-resnet-50")
model = DetrForSegmentation.from_pretrained("facebook/detr-resnet-50")
image = Image.open(<SOURCE_IMAGE_PATH>)
inputs = processor(images=image, return_tensors="pt")
@@ -79,8 +79,8 @@ size relative to the image resolution.
with torch.no_grad():
outputs = model(**inputs)
width, height = image.size
target_size = torch.tensor([[height, width]])
width, height = image_slice.size
target_size = torch.tensor([[width, height]])
results = processor.post_process_object_detection(
outputs=outputs, target_sizes=target_size)[0]
detections = sv.Detections.from_transformers(results)
@@ -175,7 +175,7 @@ objects within each, and aggregating the results.
def callback(image_slice: np.ndarray) -> sv.Detections:
results = model.infer(image_slice)[0]
detections = sv.Detections.from_inference(results)
return sv.Detections.from_inference(results)
slicer = sv.InferenceSlicer(callback = callback)
detections = slicer(image)
@@ -239,8 +239,8 @@ objects within each, and aggregating the results.
with torch.no_grad():
outputs = model(**inputs)
width, height = image.size
target_size = torch.tensor([[height, width]])
width, height = image_slice.size
target_size = torch.tensor([[width, height]])
results = processor.post_process_object_detection(
outputs=outputs, target_sizes=target_size)[0]
return sv.Detections.from_transformers(results)
@@ -264,3 +264,63 @@ objects within each, and aggregating the results.
```
![detection-with-inference-slicer](https://media.roboflow.com/supervision_detect_small_objects_example_3.png)
## Small Object Segmentation
[`InferenceSlicer`](/latest/detection/tools/inference_slicer/#supervision.detection.tools.inference_slicer.InferenceSlicer) can perform segmentation tasks too.
=== "Inference"
```{ .py hl_lines="6 16 19-20" }
import cv2
import numpy as np
import supervision as sv
from inference import get_model
model = get_model(model_id="yolov8x-seg-640")
image = cv2.imread(<SOURCE_IMAGE_PATH>)
def callback(image_slice: np.ndarray) -> sv.Detections:
results = model.infer(image_slice)[0]
detections = sv.Detections.from_inference(results)
slicer = sv.InferenceSlicer(callback = callback)
detections = slicer(image)
mask_annotator = sv.MaskAnnotator()
label_annotator = sv.LabelAnnotator()
annotated_image = mask_annotator.annotate(
scene=image, detections=detections)
annotated_image = label_annotator.annotate(
scene=annotated_image, detections=detections)
```
=== "Ultralytics"
```{ .py hl_lines="6 16 19-20" }
import cv2
import numpy as np
import supervision as sv
from ultralytics import YOLO
model = YOLO("yolov8x-seg.pt")
image = cv2.imread(<SOURCE_IMAGE_PATH>)
def callback(image_slice: np.ndarray) -> sv.Detections:
result = model(image_slice)[0]
return sv.Detections.from_ultralytics(result)
slicer = sv.InferenceSlicer(callback = callback)
detections = slicer(image)
mask_annotator = sv.MaskAnnotator()
label_annotator = sv.LabelAnnotator()
annotated_image = mask_annotator.annotate(
scene=image, detections=detections)
annotated_image = label_annotator.annotate(
scene=annotated_image, detections=detections)
```
![detection-with-inference-slicer](https://media.roboflow.com/supervision-docs/inference-slicer-segmentation-example.png)
+39 -2
View File
@@ -13,7 +13,10 @@ status: new
image = ...
key_points = sv.KeyPoints(...)
vertex_annotator = sv.VertexAnnotator(color=sv.Color.GREEN, radius=10)
vertex_annotator = sv.VertexAnnotator(
color=sv.Color.GREEN,
radius=10
)
annotated_frame = vertex_annotator.annotate(
scene=image.copy(),
key_points=key_points
@@ -34,7 +37,10 @@ status: new
image = ...
key_points = sv.KeyPoints(...)
edge_annotator = sv.EdgeAnnotator(color=sv.Color.GREEN, thickness=5)
edge_annotator = sv.EdgeAnnotator(
color=sv.Color.GREEN,
thickness=5
)
annotated_frame = edge_annotator.annotate(
scene=image.copy(),
key_points=key_points
@@ -47,6 +53,31 @@ status: new
</div>
=== "VertexLabelAnnotator"
```python
import supervision as sv
image = ...
key_points = sv.KeyPoints(...)
vertex_label_annotator = sv.VertexLabelAnnotator(
color=sv.Color.GREEN,
text_color=sv.Color.BLACK,
border_radius=5
)
annotated_frame = vertex_label_annotator.annotate(
scene=image.copy(),
key_points=key_points
)
```
<div class="result" markdown>
![vertex-label-annotator-example](https://media.roboflow.com/supervision-annotator-examples/vertex-label-annotator-example.png){ align=center width="800" }
</div>
<div class="md-typeset">
<h2><a href="#supervision.keypoint.annotators.VertexAnnotator">VertexAnnotator</a></h2>
</div>
@@ -58,3 +89,9 @@ status: new
</div>
:::supervision.keypoint.annotators.EdgeAnnotator
<div class="md-typeset">
<h2><a href="#supervision.keypoint.annotators.VertexLabelAnnotator">VertexLabelAnnotator</a></h2>
</div>
:::supervision.keypoint.annotators.VertexLabelAnnotator
Generated
+21 -21
View File
@@ -2198,13 +2198,13 @@ pygments = ">2.12.0"
[[package]]
name = "mkdocs-material"
version = "9.5.21"
version = "9.5.22"
description = "Documentation that simply works"
optional = false
python-versions = ">=3.8"
files = [
{file = "mkdocs_material-9.5.21-py3-none-any.whl", hash = "sha256:210e1f179682cd4be17d5c641b2f4559574b9dea2f589c3f0e7c17c5bd1959bc"},
{file = "mkdocs_material-9.5.21.tar.gz", hash = "sha256:049f82770f40559d3c2aa2259c562ea7257dbb4aaa9624323b5ef27b2d95a450"},
{file = "mkdocs_material-9.5.22-py3-none-any.whl", hash = "sha256:8c7a377d323567934e6cd46915e64dc209efceaec0dec1cf2202184f5649862c"},
{file = "mkdocs_material-9.5.22.tar.gz", hash = "sha256:22a853a456ae8c581c4628159574d6fc7c71b2c7569dc9c3a82cc70432219599"},
]
[package.dependencies]
@@ -3661,28 +3661,28 @@ files = [
[[package]]
name = "ruff"
version = "0.4.3"
version = "0.4.4"
description = "An extremely fast Python linter and code formatter, written in Rust."
optional = false
python-versions = ">=3.7"
files = [
{file = "ruff-0.4.3-py3-none-macosx_10_12_x86_64.whl", hash = "sha256:b70800c290f14ae6fcbb41bbe201cf62dfca024d124a1f373e76371a007454ce"},
{file = "ruff-0.4.3-py3-none-macosx_11_0_arm64.whl", hash = "sha256:08a0d6a22918ab2552ace96adeaca308833873a4d7d1d587bb1d37bae8728eb3"},
{file = "ruff-0.4.3-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:eba1f14df3c758dd7de5b55fbae7e1c8af238597961e5fb628f3de446c3c40c5"},
{file = "ruff-0.4.3-py3-none-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:819fb06d535cc76dfddbfe8d3068ff602ddeb40e3eacbc90e0d1272bb8d97113"},
{file = "ruff-0.4.3-py3-none-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0bfc9e955e6dc6359eb6f82ea150c4f4e82b660e5b58d9a20a0e42ec3bb6342b"},
{file = "ruff-0.4.3-py3-none-manylinux_2_17_ppc64.manylinux2014_ppc64.whl", hash = "sha256:510a67d232d2ebe983fddea324dbf9d69b71c4d2dfeb8a862f4a127536dd4cfb"},
{file = "ruff-0.4.3-py3-none-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:dc9ff11cd9a092ee7680a56d21f302bdda14327772cd870d806610a3503d001f"},
{file = "ruff-0.4.3-py3-none-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:29efff25bf9ee685c2c8390563a5b5c006a3fee5230d28ea39f4f75f9d0b6f2f"},
{file = "ruff-0.4.3-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:18b00e0bcccf0fc8d7186ed21e311dffd19761cb632241a6e4fe4477cc80ef6e"},
{file = "ruff-0.4.3-py3-none-musllinux_1_2_aarch64.whl", hash = "sha256:262f5635e2c74d80b7507fbc2fac28fe0d4fef26373bbc62039526f7722bca1b"},
{file = "ruff-0.4.3-py3-none-musllinux_1_2_armv7l.whl", hash = "sha256:7363691198719c26459e08cc17c6a3dac6f592e9ea3d2fa772f4e561b5fe82a3"},
{file = "ruff-0.4.3-py3-none-musllinux_1_2_i686.whl", hash = "sha256:eeb039f8428fcb6725bb63cbae92ad67b0559e68b5d80f840f11914afd8ddf7f"},
{file = "ruff-0.4.3-py3-none-musllinux_1_2_x86_64.whl", hash = "sha256:927b11c1e4d0727ce1a729eace61cee88a334623ec424c0b1c8fe3e5f9d3c865"},
{file = "ruff-0.4.3-py3-none-win32.whl", hash = "sha256:25cacda2155778beb0d064e0ec5a3944dcca9c12715f7c4634fd9d93ac33fd30"},
{file = "ruff-0.4.3-py3-none-win_amd64.whl", hash = "sha256:7a1c3a450bc6539ef00da6c819fb1b76b6b065dec585f91456e7c0d6a0bbc725"},
{file = "ruff-0.4.3-py3-none-win_arm64.whl", hash = "sha256:71ca5f8ccf1121b95a59649482470c5601c60a416bf189d553955b0338e34614"},
{file = "ruff-0.4.3.tar.gz", hash = "sha256:ff0a3ef2e3c4b6d133fbedcf9586abfbe38d076041f2dc18ffb2c7e0485d5a07"},
{file = "ruff-0.4.4-py3-none-macosx_10_12_x86_64.whl", hash = "sha256:29d44ef5bb6a08e235c8249294fa8d431adc1426bfda99ed493119e6f9ea1bf6"},
{file = "ruff-0.4.4-py3-none-macosx_11_0_arm64.whl", hash = "sha256:c4efe62b5bbb24178c950732ddd40712b878a9b96b1d02b0ff0b08a090cbd891"},
{file = "ruff-0.4.4-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4c8e2f1e8fc12d07ab521a9005d68a969e167b589cbcaee354cb61e9d9de9c15"},
{file = "ruff-0.4.4-py3-none-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:60ed88b636a463214905c002fa3eaab19795679ed55529f91e488db3fe8976ab"},
{file = "ruff-0.4.4-py3-none-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b90fc5e170fc71c712cc4d9ab0e24ea505c6a9e4ebf346787a67e691dfb72e85"},
{file = "ruff-0.4.4-py3-none-manylinux_2_17_ppc64.manylinux2014_ppc64.whl", hash = "sha256:8e7e6ebc10ef16dcdc77fd5557ee60647512b400e4a60bdc4849468f076f6eef"},
{file = "ruff-0.4.4-py3-none-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b9ddb2c494fb79fc208cd15ffe08f32b7682519e067413dbaf5f4b01a6087bcd"},
{file = "ruff-0.4.4-py3-none-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c51c928a14f9f0a871082603e25a1588059b7e08a920f2f9fa7157b5bf08cfe9"},
{file = "ruff-0.4.4-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b5eb0a4bfd6400b7d07c09a7725e1a98c3b838be557fee229ac0f84d9aa49c36"},
{file = "ruff-0.4.4-py3-none-musllinux_1_2_aarch64.whl", hash = "sha256:b1867ee9bf3acc21778dcb293db504692eda5f7a11a6e6cc40890182a9f9e595"},
{file = "ruff-0.4.4-py3-none-musllinux_1_2_armv7l.whl", hash = "sha256:1aecced1269481ef2894cc495647392a34b0bf3e28ff53ed95a385b13aa45768"},
{file = "ruff-0.4.4-py3-none-musllinux_1_2_i686.whl", hash = "sha256:9da73eb616b3241a307b837f32756dc20a0b07e2bcb694fec73699c93d04a69e"},
{file = "ruff-0.4.4-py3-none-musllinux_1_2_x86_64.whl", hash = "sha256:958b4ea5589706a81065e2a776237de2ecc3e763342e5cc8e02a4a4d8a5e6f95"},
{file = "ruff-0.4.4-py3-none-win32.whl", hash = "sha256:cb53473849f011bca6e754f2cdf47cafc9c4f4ff4570003a0dad0b9b6890e876"},
{file = "ruff-0.4.4-py3-none-win_amd64.whl", hash = "sha256:424e5b72597482543b684c11def82669cc6b395aa8cc69acc1858b5ef3e5daae"},
{file = "ruff-0.4.4-py3-none-win_arm64.whl", hash = "sha256:39df0537b47d3b597293edbb95baf54ff5b49589eb7ff41926d8243caa995ea6"},
{file = "ruff-0.4.4.tar.gz", hash = "sha256:f87ea42d5cdebdc6a69761a9d0bc83ae9b3b30d0ad78952005ba6568d6c022af"},
]
[[package]]
+1 -1
View File
@@ -1,6 +1,6 @@
[tool.poetry]
name = "supervision"
version = "0.21.0rc4"
version = "0.21.0rc5"
description = "A set of easy-to-use utils that will come in handy in any Computer Vision project"
authors = ["Piotr Skalski <piotr.skalski92@gmail.com>"]
maintainers = ["Piotr Skalski <piotr.skalski92@gmail.com>"]
+8 -1
View File
@@ -46,12 +46,15 @@ from supervision.detection.utils import (
box_iou_batch,
box_non_max_suppression,
calculate_masks_centroids,
clip_boxes,
filter_polygons_by_area,
mask_iou_batch,
mask_non_max_suppression,
mask_to_polygons,
mask_to_xyxy,
move_boxes,
move_masks,
pad_boxes,
polygon_to_mask,
polygon_to_xyxy,
scale_boxes,
@@ -69,7 +72,11 @@ from supervision.draw.utils import (
)
from supervision.geometry.core import Point, Position, Rect
from supervision.geometry.utils import get_polygon_center
from supervision.keypoint.annotators import EdgeAnnotator, VertexAnnotator
from supervision.keypoint.annotators import (
EdgeAnnotator,
VertexAnnotator,
VertexLabelAnnotator,
)
from supervision.keypoint.core import KeyPoints
from supervision.metrics.detection import ConfusionMatrix, MeanAveragePrecision
from supervision.tracker.byte_tracker.core import ByteTrack
@@ -4,20 +4,36 @@ from typing import Callable, Optional, Tuple
import numpy as np
from supervision.detection.core import Detections
from supervision.detection.utils import move_boxes
from supervision.detection.utils import move_boxes, move_masks
from supervision.utils.image import crop_image
def move_detections(detections: Detections, offset: np.array) -> Detections:
def move_detections(
detections: Detections,
offset: np.ndarray,
resolution_wh: Optional[Tuple[int, int]] = None,
) -> Detections:
"""
Args:
detections (sv.Detections): Detections object to be moved.
offset (np.array): An array of shape `(2,)` containing offset values in format
offset (np.ndarray): An array of shape `(2,)` containing offset values in format
is `[dx, dy]`.
resolution_wh (Tuple[int, int]): The width and height of the desired mask
resolution. Required for segmentation detections.
Returns:
(sv.Detections) repositioned Detections object.
"""
detections.xyxy = move_boxes(xyxy=detections.xyxy, offset=offset)
if detections.mask is not None:
if resolution_wh is None:
raise ValueError(
"Resolution width and height are required for moving segmentation "
"detections. This should be the same as (width, height) of image shape."
)
detections.mask = move_masks(
masks=detections.mask, offset=offset, resolution_wh=resolution_wh
)
return detections
@@ -126,7 +142,10 @@ class InferenceSlicer:
"""
image_slice = crop_image(image=image, xyxy=offset)
detections = self.callback(image_slice)
detections = move_detections(detections=detections, offset=offset[:2])
resolution_wh = (image.shape[1], image.shape[0])
detections = move_detections(
detections=detections, offset=offset[:2], resolution_wh=resolution_wh
)
return detections
+64 -1
View File
@@ -297,6 +297,35 @@ def clip_boxes(xyxy: np.ndarray, resolution_wh: Tuple[int, int]) -> np.ndarray:
return result
def pad_boxes(xyxy: np.ndarray, px: int, py: Optional[int] = None) -> np.ndarray:
"""
Pads bounding boxes coordinates with a constant padding.
Args:
xyxy (np.ndarray): A numpy array of shape `(N, 4)` where each
row corresponds to a bounding box in the format
`(x_min, y_min, x_max, y_max)`.
px (int): The padding value to be added to both the left and right sides of
each bounding box.
py (Optional[int]): The padding value to be added to both the top and bottom
sides of each bounding box. If not provided, `px` will be used for both
dimensions.
Returns:
np.ndarray: A numpy array of shape `(N, 4)` where each row corresponds to a
bounding box with coordinates padded according to the provided padding
values.
"""
if py is None:
py = px
result = xyxy.copy()
result[:, [0, 1]] -= [px, py]
result[:, [2, 3]] += [px, py]
return result
def xywh_to_xyxy(boxes_xywh: np.ndarray) -> np.ndarray:
xyxy = boxes_xywh.copy()
xyxy[:, 2] = boxes_xywh[:, 0] + boxes_xywh[:, 2]
@@ -500,7 +529,7 @@ def process_roboflow_result(
np.ndarray,
Optional[np.ndarray],
Optional[np.ndarray],
Dict[str, List[np.ndarray]],
Dict[str, Union[List[np.ndarray], np.ndarray]],
]:
if not roboflow_result["predictions"]:
return (
@@ -592,6 +621,40 @@ def move_boxes(xyxy: np.ndarray, offset: np.ndarray) -> np.ndarray:
return xyxy + np.hstack([offset, offset])
def move_masks(
masks: np.ndarray,
offset: np.ndarray,
resolution_wh: Tuple[int, int] = None,
) -> np.ndarray:
"""
Offset the masks in an array by the specified (x, y) amount.
Args:
masks (np.ndarray): A 3D array of binary masks corresponding to the predictions.
Shape: `(N, H, W)`, where N is the number of predictions, and H, W are the
dimensions of each mask.
offset (np.ndarray): An array of shape `(2,)` containing non-negative int values
`[dx, dy]`.
resolution_wh (Tuple[int, int]): The width and height of the desired mask
resolution.
Returns:
(np.ndarray) repositioned masks, optionally padded to the specified shape.
"""
if offset[0] < 0 or offset[1] < 0:
raise ValueError(f"Offset values must be non-negative integers. Got: {offset}")
mask_array = np.full((masks.shape[0], resolution_wh[1], resolution_wh[0]), False)
mask_array[
:,
offset[1] : masks.shape[1] + offset[1],
offset[0] : masks.shape[2] + offset[0],
] = masks
return mask_array
def scale_boxes(xyxy: np.ndarray, factor: float) -> np.ndarray:
"""
Scale the dimensions of bounding boxes.
+52
View File
@@ -81,6 +81,58 @@ def draw_filled_rectangle(scene: np.ndarray, rect: Rect, color: Color) -> np.nda
return scene
def draw_rounded_rectangle(
scene: np.ndarray,
rect: Rect,
color: Color,
border_radius: int,
) -> np.ndarray:
"""
Draws a rounded rectangle on an image.
Parameters:
scene (np.ndarray): The image on which the rounded rectangle will be drawn.
rect (Rect): The rectangle to be drawn.
color (Color): The color of the rounded rectangle.
border_radius (int): The radius of the corner rounding.
Returns:
np.ndarray: The image with the rounded rectangle drawn on it.
"""
x1, y1, x2, y2 = rect.as_xyxy_int_tuple()
width, height = x2 - x1, y2 - y1
border_radius = min(border_radius, min(width, height) // 2)
rectangle_coordinates = [
((x1 + border_radius, y1), (x2 - border_radius, y2)),
((x1, y1 + border_radius), (x2, y2 - border_radius)),
]
circle_centers = [
(x1 + border_radius, y1 + border_radius),
(x2 - border_radius, y1 + border_radius),
(x1 + border_radius, y2 - border_radius),
(x2 - border_radius, y2 - border_radius),
]
for coordinates in rectangle_coordinates:
cv2.rectangle(
img=scene,
pt1=coordinates[0],
pt2=coordinates[1],
color=color.as_bgr(),
thickness=-1,
)
for center in circle_centers:
cv2.circle(
img=scene,
center=center,
radius=border_radius,
color=color.as_bgr(),
thickness=-1,
)
return scene
def draw_polygon(
scene: np.ndarray, polygon: np.ndarray, color: Color, thickness: int = 2
) -> np.ndarray:
+13
View File
@@ -98,6 +98,11 @@ class Rect:
width: float
height: float
@classmethod
def from_xyxy(cls, xyxy: Tuple[float, float, float, float]) -> Rect:
x1, y1, x2, y2 = xyxy
return cls(x=x1, y=y1, width=x2 - x1, height=y2 - y1)
@property
def top_left(self) -> Point:
return Point(x=self.x, y=self.y)
@@ -113,3 +118,11 @@ class Rect:
width=self.width + 2 * padding,
height=self.height + 2 * padding,
)
def as_xyxy_int_tuple(self) -> Tuple[int, int, int, int]:
return (
int(self.x),
int(self.y),
int(self.x + self.width),
int(self.y + self.height),
)
+247 -6
View File
@@ -1,12 +1,14 @@
from abc import ABC, abstractmethod
from logging import warn
from typing import List, Optional, Tuple
from typing import List, Optional, Tuple, Union
import cv2
import numpy as np
from supervision import Rect, pad_boxes
from supervision.annotators.base import ImageType
from supervision.draw.color import Color
from supervision.draw.utils import draw_rounded_rectangle
from supervision.keypoint.core import KeyPoints
from supervision.keypoint.skeletons import SKELETONS_BY_VERTEX_COUNT
from supervision.utils.conversion import convert_for_annotation_method
@@ -46,8 +48,8 @@ class VertexAnnotator(BaseKeyPointAnnotator):
points. It draws circles at each key point location.
Args:
scene (ImageType): The image where bounding boxes will be drawn. `ImageType`
is a flexible type, accepting either `numpy.ndarray` or
scene (ImageType): The image where skeleton vertices will be drawn.
`ImageType` is a flexible type, accepting either `numpy.ndarray` or
`PIL.Image.Image`.
key_points (KeyPoints): A collection of key points where each key point
consists of x and y coordinates.
@@ -63,7 +65,10 @@ class VertexAnnotator(BaseKeyPointAnnotator):
image = ...
key_points = sv.KeyPoints(...)
vertex_annotator = sv.VertexAnnotator(color=sv.Color.GREEN, radius=10)
vertex_annotator = sv.VertexAnnotator(
color=sv.Color.GREEN,
radius=10
)
annotated_frame = vertex_annotator.annotate(
scene=image.copy(),
key_points=key_points
@@ -119,7 +124,7 @@ class EdgeAnnotator(BaseKeyPointAnnotator):
edges.
Args:
scene (ImageType): The image where bounding boxes will be drawn. `ImageType`
scene (ImageType): The image where skeleton edges will be drawn. `ImageType`
is a flexible type, accepting either `numpy.ndarray` or
`PIL.Image.Image`.
key_points (KeyPoints): A collection of key points where each key point
@@ -137,7 +142,10 @@ class EdgeAnnotator(BaseKeyPointAnnotator):
image = ...
key_points = sv.KeyPoints(...)
edge_annotator = sv.EdgeAnnotator(color=sv.Color.GREEN, thickness=5)
edge_annotator = sv.EdgeAnnotator(
color=sv.Color.GREEN,
thickness=5
)
annotated_frame = edge_annotator.annotate(
scene=image.copy(),
key_points=key_points
@@ -175,3 +183,236 @@ class EdgeAnnotator(BaseKeyPointAnnotator):
)
return scene
class VertexLabelAnnotator:
"""
A class that draws labels of skeleton vertices on images. It uses specified key
points to determine the locations where the vertices should be drawn.
"""
def __init__(
self,
color: Union[Color, List[Color]] = Color.ROBOFLOW,
text_color: Color = Color.WHITE,
text_scale: float = 0.5,
text_thickness: int = 1,
text_padding: int = 10,
border_radius: int = 0,
):
"""
Args:
color (Union[Color, List[Color]], optional): The color to use for each
keypoint label. If a list is provided, the colors will be used in order
for each keypoint.
text_color (Color, optional): The color to use for the labels.
text_scale (float, optional): The scale of the text.
text_thickness (int, optional): The thickness of the text.
text_padding (int, optional): The padding around the text.
border_radius (int, optional): The radius of the rounded corners of the
boxes. Set to a high value to produce circles.
"""
self.border_radius: int = border_radius
self.color: Union[Color, List[Color]] = color
self.text_color: Color = text_color
self.text_scale: float = text_scale
self.text_thickness: int = text_thickness
self.text_padding: int = text_padding
def annotate(
self, scene: ImageType, key_points: KeyPoints, labels: List[str] = None
) -> ImageType:
"""
A class that draws labels of skeleton vertices on images. It uses specified key
points to determine the locations where the vertices should be drawn.
Args:
scene (ImageType): The image where vertex labels will be drawn. `ImageType`
is a flexible type, accepting either `numpy.ndarray` or
`PIL.Image.Image`.
key_points (KeyPoints): A collection of key points where each key point
consists of x and y coordinates.
labels (List[str], optional): A list of labels to be displayed on the
annotated image. If not provided, keypoint indices will be used.
Returns:
The annotated image, matching the type of `scene` (`numpy.ndarray`
or `PIL.Image.Image`)
Example:
```python
import supervision as sv
image = ...
key_points = sv.KeyPoints(...)
vertex_label_annotator = sv.VertexLabelAnnotator(
color=sv.Color.GREEN,
text_color=sv.Color.BLACK,
border_radius=5
)
annotated_frame = vertex_label_annotator.annotate(
scene=image.copy(),
key_points=key_points
)
```
![vertex-label-annotator-example](https://media.roboflow.com/
supervision-annotator-examples/vertex-label-annotator-example.png)
!!! tip
`VertexLabelAnnotator` allows to customize the color of each keypoint label
values.
Example:
```python
import supervision as sv
image = ...
key_points = sv.KeyPoints(...)
LABELS = [
"nose", "left eye", "right eye", "left ear",
"right ear", "left shoulder", "right shoulder", "left elbow",
"right elbow", "left wrist", "right wrist", "left hip",
"right hip", "left knee", "right knee", "left ankle",
"right ankle"
]
COLORS = [
"#FF6347", "#FF6347", "#FF6347", "#FF6347",
"#FF6347", "#FF1493", "#00FF00", "#FF1493",
"#00FF00", "#FF1493", "#00FF00", "#FFD700",
"#00BFFF", "#FFD700", "#00BFFF", "#FFD700",
"#00BFFF"
]
COLORS = [sv.Color.from_hex(color_hex=c) for c in COLORS]
vertex_label_annotator = sv.VertexLabelAnnotator(
color=COLORS,
text_color=sv.Color.BLACK,
border_radius=5
)
annotated_frame = vertex_label_annotator.annotate(
scene=image.copy(),
key_points=key_points,
labels=labels
)
```
![vertex-label-annotator-custom-example](https://media.roboflow.com/
supervision-annotator-examples/vertex-label-annotator-custom-example.png)
"""
font = cv2.FONT_HERSHEY_SIMPLEX
skeletons_count, points_count, _ = key_points.xy.shape
if skeletons_count == 0:
return scene
anchors = key_points.xy.reshape(points_count * skeletons_count, 2).astype(int)
mask = np.all(anchors != 0, axis=1)
if not np.any(mask):
return scene
colors = self.preprocess_and_validate_colors(
colors=self.color,
points_count=points_count,
skeletons_count=skeletons_count,
)
labels = self.preprocess_and_validate_labels(
labels=labels, points_count=points_count, skeletons_count=skeletons_count
)
anchors = anchors[mask]
colors = colors[mask]
labels = labels[mask]
xyxy = np.array(
[
self.get_text_bounding_box(
text=label,
font=font,
text_scale=self.text_scale,
text_thickness=self.text_thickness,
center_coordinates=tuple(anchor),
)
for anchor, label in zip(anchors, labels)
]
)
xyxy_padded = pad_boxes(xyxy=xyxy, px=self.text_padding)
for text, color, box, box_padded in zip(labels, colors, xyxy, xyxy_padded):
draw_rounded_rectangle(
scene=scene,
rect=Rect.from_xyxy(box_padded),
color=color,
border_radius=self.border_radius,
)
cv2.putText(
img=scene,
text=text,
org=(box[0], box[1] + self.text_padding),
fontFace=font,
fontScale=self.text_scale,
color=self.text_color.as_rgb(),
thickness=self.text_thickness,
lineType=cv2.LINE_AA,
)
return scene
@staticmethod
def get_text_bounding_box(
text: str,
font: int,
text_scale: float,
text_thickness: int,
center_coordinates: Tuple[int, int],
) -> Tuple[int, int, int, int]:
text_w, text_h = cv2.getTextSize(
text=text,
fontFace=font,
fontScale=text_scale,
thickness=text_thickness,
)[0]
center_x, center_y = center_coordinates
return (
center_x - text_w // 2,
center_y - text_h // 2,
center_x + text_w // 2,
center_y + text_h // 2,
)
@staticmethod
def preprocess_and_validate_labels(
labels: Optional[List[str]], points_count: int, skeletons_count: int
) -> np.array:
if labels and len(labels) != points_count:
raise ValueError(
f"Number of labels ({len(labels)}) must match number of key points "
f"({points_count})."
)
if labels is None:
labels = [str(i) for i in range(points_count)]
return np.array(labels * skeletons_count)
@staticmethod
def preprocess_and_validate_colors(
colors: Optional[Union[Color, List[Color]]],
points_count: int,
skeletons_count: int,
) -> np.array:
if isinstance(colors, list) and len(colors) != points_count:
raise ValueError(
f"Number of colors ({len(colors)}) must match number of key points "
f"({points_count})."
)
return (
np.array(colors * skeletons_count)
if isinstance(colors, list)
else np.array([colors] * points_count * skeletons_count)
)