GitHub

ObjectDetector.jl

Object detection via YOLO in Julia. YOLO models are loaded directly from Darknet .cfg and .weights files as Flux models, so a Darknet architecture runs without a conversion step. Uses CUDA if available.

Supported families: v2, v2-tiny, v3, v3-spp, v3-tiny, v4, v4-tiny, v4-csp, v4-csp-x-swish, v4x-mish, v4-p5, v4-p6 (Scaled-YOLOv4), v7, v7-tiny, v7x. Other less standard models may work too.

All supported models have result parity with AlexeyAB/darknet and are tested directly against Darknet.jl.

Installation

Requires Julia 1.10+.

pkg> add ObjectDetector

For CUDA acceleration, also add and load CUDA and cuDNN.

Running a model on an image

using ObjectDetector, FileIO, ImageIO

yolomod = YOLO.v3_608_COCO(batch = 1, silent = true)

batch = emptybatch(yolomod)              # uses the GPU if one is available
img = load(joinpath(dirname(dirname(pathof(ObjectDetector))),
                    "test", "images", "dog-cycle-car.png"))

batch[:, :, :, 1], padding = prepare_image(img, yolomod)
res = yolomod(batch, detect_thresh = 0.5, overlap_thresh = 0.8)

Each column of res is one detection:

i = 1
bbox                     = res[1:4, i]
objectness_score         = res[5, i]
selected_class_confidence = res[end-2, i]
selected_class_id        = res[end-1, i]
batch_id                 = res[end, i]

Julia is column-major and Darknet is row-major, so the image matrix needs its first two dimensions permuted before going into the batch, or features come out rotated 90°. prepare_image does that conversion, along with the aspect-preserving letterbox, and returns the padding needed to map boxes back.

Non-square models load fine, but each dimension must be an integer multiple of the network's maximum stride: 32 for most models, 64 for v4_p6.

Drawing the result

imgBoxes = draw_boxes(img, yolomod, padding, res)
save("result.png", imgBoxes)

Where to go next