o W h ,@sddlmZddlZddlZddlmZddlmZddl m Z ddl m Z m Z ddlmZdd lmZdd lmZGd d d e ZGd dde ZGdddZdS))dequeN) xywh2xyxy save_one_box) TrackState) BYTETrackerSTrack)matching)GMCKalmanFilterXYWHcs~eZdZdZeZdfdd ZddZdd Zdfd d Z fd dZ e ddZ e ddZddZe ddZZS)BOTracka An extended version of the STrack class for YOLOv8, adding object tracking features. This class extends the STrack class to include additional functionalities for object tracking, such as feature smoothing, Kalman filter prediction, and reactivation of tracks. Attributes: shared_kalman (KalmanFilterXYWH): A shared Kalman filter for all instances of BOTrack. smooth_feat (np.ndarray): Smoothed feature vector. curr_feat (np.ndarray): Current feature vector. features (deque): A deque to store feature vectors with a maximum length defined by `feat_history`. alpha (float): Smoothing factor for the exponential moving average of features. mean (np.ndarray): The mean state of the Kalman filter. covariance (np.ndarray): The covariance matrix of the Kalman filter. Methods: update_features: Update features vector and smooth it using exponential moving average. predict: Predict the mean and covariance using Kalman filter. re_activate: Reactivate a track with updated features and optionally new ID. update: Update the track with new detection and frame ID. tlwh: Property that gets the current position in tlwh format `(top left x, top left y, width, height)`. multi_predict: Predict the mean and covariance of multiple object tracks using shared Kalman filter. convert_coords: Convert tlwh bounding box coordinates to xywh format. tlwh_to_xywh: Convert bounding box to xywh format `(center x, center y, width, height)`. Examples: Create a BOTrack instance and update its features >>> bo_track = BOTrack(tlwh=[100, 50, 80, 40], score=0.9, cls=1, feat=np.random.rand(128)) >>> bo_track.predict() >>> new_track = BOTrack(tlwh=[110, 60, 80, 40], score=0.85, cls=1, feat=np.random.rand(128)) >>> bo_track.update(new_track, frame_id=2) N2csFt|||d|_d|_|dur||tg|d|_d|_dS)aS Initialize a BOTrack object with temporal parameters, such as feature history, alpha, and current features. Args: tlwh (np.ndarray): Bounding box coordinates in tlwh format (top left x, top left y, width, height). score (float): Confidence score of the detection. cls (int): Class ID of the detected object. feat (np.ndarray | None): Feature vector associated with the detection. feat_history (int): Maximum length of the feature history deque. Examples: Initialize a BOTrack object with bounding box, score, class ID, and feature vector >>> tlwh = np.array([100, 50, 80, 120]) >>> score = 0.9 >>> cls = 1 >>> feat = np.random.rand(128) >>> bo_track = BOTrack(tlwh, score, cls, feat) N)maxleng?)super__init__ smooth_feat curr_featupdate_featuresrfeaturesalpha)selftlwhscoreclsfeat feat_history __class__Q/var/www/vscode/kcb/lib/python3.10/site-packages/ultralytics/trackers/bot_sort.pyr6s  zBOTrack.__init__cCsl|tj|}||_|jdur||_n|j|jd|j||_|j||jtj|j_dS)zIUpdate the feature vector and apply exponential moving average smoothing.Nr)nplinalgnormrrrrappend)rrr r r!rRs  zBOTrack.update_featurescCsB|j}|jtjkrd|d<d|d<|j||j\|_|_dS)z\Predict the object's future state using the Kalman filter to update its mean and covariance.rN)meancopystaterTracked kalman_filterpredict covariance)r mean_stater r r!r-]s  zBOTrack.predictFcs*|jdur ||jt|||dS)zHReactivate a track with updated features and optionally assign a new ID.N)rrr re_activate)r new_trackframe_idnew_idrr r!r0fs  zBOTrack.re_activatecs(|jdur ||jt||dS)zIUpdate the track with new detection information and the current frame ID.N)rrrupdate)rr1r2rr r!r4ls  zBOTrack.updatecCsJ|jdur |jS|jdd}|dd|ddd8<|S)z]Return the current bounding box position in `(top left x, top left y, width, height)` format.N)r(_tlwhr))rretr r r!rrs   z BOTrack.tlwhcCst|dkrdStdd|D}tdd|D}t|D]\}}|jtjkr6d||d<d||d<q tj ||\}}tt ||D]\}\}}|||_ |||_ qGdS)zXPredict the mean and covariance for multiple object tracks using a shared Kalman filter.rNcSsg|]}|jqSr )r(r).0str r r! sz)BOTrack.multi_predict..cSsg|]}|jqSr )r.r9r r r!r<sr&r') lenr"asarray enumerater*rr+r shared_kalman multi_predictzipr(r.)stracks multi_meanmulti_covarianceir;r(covr r r!rA{s      zBOTrack.multi_predictcCs ||S)z5Convert tlwh bounding box coordinates to xywh format.) tlwh_to_xywh)rrr r r!convert_coordss zBOTrack.convert_coordscCs2t|}|dd|ddd7<|S)zgConvert bounding box from tlwh (top-left-width-height) to xywh (center-x-center-y-width-height) format.Nr6)r"r>r))rr8r r r!rHs zBOTrack.tlwh_to_xywh)Nr)F)__name__ __module__ __qualname____doc__r r@rrr-r0r4propertyr staticmethodrArIrH __classcell__r r rr!rs!    rcsPeZdZdZdfdd ZddZddd Zd d Zd d ZfddZ Z S)BOTSORTa\ An extended version of the BYTETracker class for YOLOv8, designed for object tracking with ReID and GMC algorithm. Attributes: proximity_thresh (float): Threshold for spatial proximity (IoU) between tracks and detections. appearance_thresh (float): Threshold for appearance similarity (ReID embeddings) between tracks and detections. encoder (Any): Object to handle ReID embeddings, set to None if ReID is not enabled. gmc (GMC): An instance of the GMC algorithm for data association. args (Any): Parsed command-line arguments containing tracking parameters. Methods: get_kalmanfilter: Return an instance of KalmanFilterXYWH for object tracking. init_track: Initialize track with detections, scores, and classes. get_dists: Get distances between tracks and detections using IoU and (optionally) ReID. multi_predict: Predict and track multiple objects with YOLOv8 model. reset: Reset the BOTSORT tracker to its initial state. Examples: Initialize BOTSORT and process detections >>> bot_sort = BOTSORT(args, frame_rate=30) >>> bot_sort.init_track(dets, scores, cls, img) >>> bot_sort.multi_predict(tracks) Note: The class is designed to work with the YOLOv8 object detection model and supports ReID only if enabled via args. csft||t|jd|_|j|_|j|_|jjdkr#dd|_ dS|j r.t |j|_ dSd|_ dS)a Initialize BOTSORT object with ReID module and GMC algorithm. Args: args (object): Parsed command-line arguments containing tracking parameters. frame_rate (int): Frame rate of the video being processed. Examples: Initialize BOTSORT with command-line arguments and a specified frame rate: >>> args = parse_args() >>> bot_sort = BOTSORT(args, frame_rate=30) )methodautocSsdd|DS)NcSg|]}|qSr cpunumpyr:fr r r!r<z6BOTSORT.__init__....r )featssr r r!sz"BOTSORT.__init__..N) rrr gmc_methodgmcproximity_threshappearance_threshargsmodel with_reidReIDencoder)rrc frame_raterr r!rs   zBOTSORT.__init__cCstS)ziReturn an instance of KalmanFilterXYWH for predicting and updating object states in the tracking process.r rr r r!get_kalmanfilterszBOTSORT.get_kalmanfilterNcCs\t|dkrgS|jjr#|jdur#|||}ddt||||DSddt|||DS)zjInitialize object tracks using detection bounding boxes, scores, class labels, and optional ReID features.rNcSs"g|] \}}}}t||||qSr r)r:xyxyr]crZr r r!r<s"z&BOTSORT.init_track..cSsg|] \}}}t|||qSr rk)r:rlr]rmr r r!r<s)r=rcrergrB)rdetsscoresrimg features_keepr r r! init_tracks  zBOTSORT.init_trackcCszt||}|d|jk}|jjrt||}|jjr;|jdur;t||d}d||d|jk<d||<t ||}|S)z[Calculate distances between tracks and detections using IoU and optionally ReID embeddings.rNg@g?) r iou_distancerarc fuse_scorerergembedding_distancerbr"minimum)rtracks detectionsdists dists_mask emb_distsr r r! get_distss   zBOTSORT.get_distscCst|dS)zWPredict the mean and covariance of multiple object tracks using a shared Kalman filter.N)rrA)rrwr r r!rAszBOTSORT.multi_predictcst|jdS)zaReset the BOTSORT tracker to its initial state, clearing all tracked objects and internal states.N)rresetr` reset_paramsrirr r!r}s z BOTSORT.reset)rR)N) rJrKrLrMrrjrrr|rAr}rPr r rr!rQs  rQc@s eZdZdZddZddZdS)rfz,YOLO model as encoder for re-identification.cCsDddlm}|||_|jd|vrt|jjjdndgdddS) z)Initialize encoder for re-identification.r)YOLOz.ptr6F)embedverboseN) ultralyticsrrdr=)rrdrr r r!rs  .z ReID.__init__c s@|fddtt|ddddfD}dd|DS)z(Extract embeddings for detected objects.csg|] }t|ddqS)F)saver)r:detrpr r!r<sz!ReID.__call__..Nr5cSrUr rVrYr r r!r<r[)rdrtorch from_numpy)rrprnr\r rr!__call__s2z ReID.__call__N)rJrKrLrMrrr r r r!rfs rf) collectionsrrXr"rultralytics.utils.opsrultralytics.utils.plottingr basetrackr byte_trackerrr utilsr utils.gmcr utils.kalman_filterr rrQrfr r r r!s       ^