v0.6.0

enhancement

• add coco_evaluation script, refactor coco_error_analysis script (#162)
  coco_evaluation.py script usage:

python scripts/coco_evaluation.py dataset.json results.json

will calculate coco evaluation and export them to given output folder directory.

If you want to specify mAP metric type, set it as --metric bbox mask.

If you want to also calculate classwise scores add --classwise argument.

If you want to specify max detections, set it as --proposal_nums 10 100 500.

If you want to specify a psecific IOU threshold, set it as --iou_thrs 0.5. Default includes 0.50:0.95 and 0.5 scores.

If you want to specify export directory, set it as --out_dir output/folder/directory .

coco_error_analysis.py script usage:

python scripts/coco_error_analysis.py dataset.json results.json

will calculate coco error plots and export them to given output folder directory.

If you want to specify mAP result type, set it as --types bbox mask.

If you want to export extra mAP bar plots and annotation area stats add --extraplots argument.

If you want to specify area regions, set it as --areas 1024 9216 10000000000.

If you want to specify export directory, set it as --out_dir output/folder/directory .

bugfixes

• prevent empty bbox coco json creation (#164)
• dont create mot info while when type='det' (#163)

breaking changes

• refactor predict (#161)
  By default, scripts apply both standard and sliced prediction (multi-stage inference). If you don't want to perform sliced prediction add --no_sliced_pred argument. If you don't want to perform standard prediction add --no_standard_pred argument.

v0.5.2

• fix negative bbox coord error (#160)

v0.5.1

• add predict_fiftyone script to perform sliced/standard inference over yolov5/mmdetection models and visualize the incorrect prediction over the fiftyone ui.

• fix mot utils (#152)

v0.5.0

• add check for image size in slice_image (#147)
• refactor prediction output (#148)
• fix slice_image in readme (#149)

refactor prediction output

# perform standard or sliced prediction
result = get_prediction(image, detection_model)
result = get_sliced_prediction(image, detection_model)

# export prediction visuals to "demo_data/"
result.export_visuals(export_dir="demo_data/")

# convert predictions to coco annotations
result.to_coco_annotations()

# convert predictions to coco predictions
result.to_coco_predictions(image_id=1)

# convert predictions to [imantics](https://github.com/jsbroks/imantics) annotation format
result.to_imantics_annotations()

# convert predictions to [fiftyone](https://github.com/voxel51/fiftyone) detection format
result.to_fiftyone_detections()

• Check more in colab notebooks:

YOLOv5 + SAHI demo:

MMDetection + SAHI demo:

v0.4.8

• update mot utils (#143)
• add fiftyone utils (#144)

FiftyOne Utilities

Explore COCO dataset via FiftyOne app:

from sahi.utils.fiftyone import launch_fiftyone_app
# launch fiftyone app:
session = launch_fiftyone_app(coco_image_dir, coco_json_path)
# close fiftyone app:
session.close()

Convert predictions to FiftyOne detection:

from sahi import get_sliced_prediction
# perform sliced prediction
result = get_sliced_prediction(
    image,
    detection_model,
    slice_height = 256,
    slice_width = 256,
    overlap_height_ratio = 0.2,
    overlap_width_ratio = 0.2
)
# convert first object into fiftyone detection format
object_prediction = result["object_prediction_list"][0]
fiftyone_detection = object_prediction.to_fiftyone_detection(image_height=720, image_width=1280)

v0.4.6

new feature

• add more mot utils (#133)

MOT Challenge formatted ground truth dataset creation:

• import required classes:

from sahi.utils.mot import MotAnnotation, MotFrame, MotVideo

• init video:

mot_video = MotVideo(name="sequence_name")

• init first frame:

mot_frame = MotFrame()

• add annotations to frame:

mot_frame.add_annotation(
  MotAnnotation(bbox=[x_min, y_min, width, height])
)

mot_frame.add_annotation(
  MotAnnotation(bbox=[x_min, y_min, width, height])
)

• add frame to video:

mot_video.add_frame(mot_frame)

• export in MOT challenge format:

mot_video.export(export_dir="mot_gt", type="gt")

• your MOT challenge formatted ground truth files are ready under mot_gt/sequence_name/ folder.

Advanced MOT Challenge formatted ground truth dataset creation:

• you can customize tracker while initializing mot video object:

tracker_params = {
  'max_distance_between_points': 30,
  'min_detection_threshold': 0,
  'hit_inertia_min': 10,
  'hit_inertia_max': 12,
  'point_transience': 4,
}
# for details: https://github.com/tryolabs/norfair/tree/master/docs#arguments

mot_video = MotVideo(tracker_kwargs=tracker_params)

• you can omit automatic track id generation and directly provide track ids of annotations:

# create annotations with track ids:
mot_frame.add_annotation(
  MotAnnotation(bbox=[x_min, y_min, width, height], track_id=1)
)

mot_frame.add_annotation(
  MotAnnotation(bbox=[x_min, y_min, width, height], track_id=2)
)

# add frame to video:
mot_video.add_frame(mot_frame)

# export in MOT challenge format without automatic track id generation:
mot_video.export(export_dir="mot_gt", type="gt", use_tracker=False)

• you can overwrite the results into already present directory by adding exist_ok=True:

mot_video.export(export_dir="mot_gt", type="gt", exist_ok=True)

MOT Challenge formatted tracker output creation:

• import required classes:

from sahi.utils.mot import MotAnnotation, MotFrame, MotVideo

• init video by providing video name:

mot_video = MotVideo(name="sequence_name")

• init first frame:

mot_frame = MotFrame()

• add tracker outputs to frame:

mot_frame.add_annotation(
  MotAnnotation(bbox=[x_min, y_min, width, height], track_id=1)
)

mot_frame.add_annotation(
  MotAnnotation(bbox=[x_min, y_min, width, height], track_id=2)
)

• add frame to video:

mot_video.add_frame(mot_frame)

• export in MOT challenge format:

mot_video.export(export_dir="mot_test", type="test")

• your MOT challenge formatted ground truth files are ready as mot_test/sequence_name.txt.

Advanced MOT Challenge formatted tracker output creation:

• you can enable tracker and directly provide object detector output:

# add object detector outputs:
mot_frame.add_annotation(
  MotAnnotation(bbox=[x_min, y_min, width, height])
)

mot_frame.add_annotation(
  MotAnnotation(bbox=[x_min, y_min, width, height])
)

# add frame to video:
mot_video.add_frame(mot_frame)

# export in MOT challenge format by applying a kalman based tracker:
mot_video.export(export_dir="mot_gt", type="gt", use_tracker=True)

• you can customize tracker while initializing mot video object:

tracker_params = {
  'max_distance_between_points': 30,
  'min_detection_threshold': 0,
  'hit_inertia_min': 10,
  'hit_inertia_max': 12,
  'point_transience': 4,
}
# for details: https://github.com/tryolabs/norfair/tree/master/docs#arguments

mot_video = MotVideo(tracker_kwargs=tracker_params)

• you can overwrite the results into already present directory by adding exist_ok=True:

mot_video.export(export_dir="mot_gt", type="gt", exist_ok=True)

documentation

• update coco docs (#134)
• add colab links into readme (#135)

Check YOLOv5 + SAHI demo:

Check MMDetection + SAHI demo:

bug fixes

• fix demo notebooks (#136)

v0.4.5

enhancement

• add colab demo support (#127)
• add warning for image files without suffix (#129)
• seperate mmdet/yolov5 utils (#130)

v0.4.4

new feature

• add mot utils (#122)
  read mot utils doc for more detail

documentation

• update installation (#118)
• add details for coco2yolov5 usage (#120)

bug fixes

• fix typo (#117)
• update coco2yolov5.py (#115)

breaking changes

• drop python 3.6 support (#123)

v0.4.3

refactorize postprocess (#109)

• specify postprocess type as --postprocess_type UNIONMERGE or --postprocess_type NMS to be applied over sliced predictions
• specify postprocess match metric as --match_metric IOS for intersection over smaller area or --match_metric IOU for intersection over union
• specify postprocess match threshold as --match_thresh 0.5
• add --class_agnostic argument to ignore category ids of the predictions during postprocess (merging/nms)

export visuals with gt (#107)

• export visuals with predicted + gt annotations into visuals_with_gt folder when coco_file_path is provided
• keep source folder structure when exporting results
• add from_coco_annotation_dict classmethod to ObjectAnnotation
• remove unused imports/classes/parameters
• better typing hints

v0.4.2

• add yolov5 support: check yolov5 demo notebook

CLI usage:

python scripts/predict.py --model_type yolov5 --source image/file/or/folder --model_path path/to/model