Darknet Convolution2D followed by BatchNormalization and LeakyReLU.

Darknet Convolution2D followed by BatchNormalization and LeakyReLU.

Compose arbitrarily many functions, evaluated left to right. Reference: https://mathieularose.com/function-composition-in-python/

Wrapper to set Darknet parameters for Convolution2D.

Returns True if this track is confirmed.

A series of resblocks starting with a downsampling Convolution2D

Perform Kalman filter measurement update step and update the feature cache. Parameters ---------- kf : kalman_filter.

Convert bounding box to format `(center x, center y, aspect ratio, height)`, where the aspect ratio is `width / height`.

Convert final layer features to bounding box parameters.

Return iou tensor Parameters ---------- b1: tensor, shape=(i1,...,iN, 4), xywh b2: tensor, shape=(j, 4), xywh Returns -

Project state distribution to measurement space. Parameters ---------- mean : ndarray The state's mean vector (8

Compute sigmoid focal loss. Reference Paper: "Focal Loss for Dense Object Detection" https://arxiv.org/abs/1708.02002 #

Get current position in bounding box format `(min x, miny, max x, max y)`. Returns ------- ndarray The bo

Get current position in bounding box format `(top left x, top left y, width, height)`. Returns ------- ndarray

Evaluate YOLO model on given input and return filtered boxes.

Compute pair-wise cosine distance between points in `a` and `b`. Parameters ---------- a : array_like An NxM matrix of N samples

resize image with unchanged aspect ratio using padding

Compute pair-wise squared distance between points in `a` and `b`. Parameters ---------- a : array_like An NxM matrix of N samples

Calculate DIoU loss on anchor boxes Reference Paper: "Distance-IoU Loss: Faster and Better Learning for Bounding Box Regression"

Calculate GIoU loss on anchor boxes Reference Paper: "Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regre

Darknent body having 52 Convolution2D layers

Compute distance between features and targets. Parameters ---------- features : ndarray An NxM matrix of N featur

Extract image patch from bounding box. Parameters ---------- image : ndarray The full image. bbox : array_like The bo

Compute gating distance between state distribution and measurements. A suitable distance threshold can be obtained from `chi2inv95`. If

Generate detections with features. Parameters ---------- encoder : Callable[image, ndarray] -> ndarray The encoder function takes

Create track from unassociated measurement. Parameters ---------- measurement : ndarray Bounding box coordinates

Computer intersection over union. Parameters ---------- bbox : ndarray A bounding box in format `(top left x, top left y, width,

Returns True if this track is dead and should be deleted.

Mark this track as missed (no association at the current time step).

Solve linear assignment problem. Parameters ---------- distance_metric : Callable[List[Track], List[Detection], List[int], List[int]) ->

Parse command line arguments.

Update the distance metric with new data. Parameters ---------- features : ndarray An NxM matrix of N features of

Propagate the state distribution to the current time step using a Kalman filter prediction step. Parameters ----------

Compute softmax focal loss. Reference Paper: "Focal Loss for Dense Object Detection" https://arxiv.org/abs/1708.02002 #

Create YOLO_V4 model CNN body in Keras.

Process Conv layer output

Get corrected boxes

Helper function for nearest neighbor distance metric (cosine). Parameters ---------- x : ndarray A matrix of N row-vectors (samp

Helper function for nearest neighbor distance metric (Euclidean). Parameters ---------- x : ndarray A matrix of N row-vectors (s

Invalidate infeasible entries in cost matrix based on the state distributions obtained by Kalman filtering. Parameters ---------- kf

An intersection over union distance metric. Parameters ---------- tracks : List[deep_sort.track.Track] A list of tracks. dete

Returns True if this track is tentative (unconfirmed).

6 Conv2D_BN_Leaky layers followed by a Conv2D_linear layer

Run matching cascade. Parameters ---------- distance_metric : Callable[List[Track], List[Detection], List[int], List[int]) -> ndarray

Suppress overlapping detections. Original code from [1]_ has been adapted to include confidence score. .. [1] http://www.pyimagesearch.com/2

Run Kalman filter prediction step. Parameters ---------- mean : ndarray The 8 dimensional mean vector of the obje

Propagate track state distributions one time step forward. This function should be called once every time step, before `update`.

Preprocess true boxes to training input format Parameters ---------- true_boxes: array, shape=(m, T, 5) Absolute x_min, y_min, x_

Convert bounding box to format `(min x, min y, max x, max y)`, i.e., `(top left, bottom right)`.

Run Kalman filter correction step. Parameters ---------- mean : ndarray The predicted state's mean vector (8 dime

Perform measurement update and track management. Parameters ---------- detections : List[deep_sort.detection.Detection]

Return yolo4_loss tensor Parameters ---------- yolo_outputs: list of tensor, the output of yolo_body or tiny_yolo_body y_true: list o

Return yolo_loss tensor Parameters ---------- yolo_outputs: list of tensor, the output of yolo_body or tiny_yolo_body y_true: list of