FPhF^ddlZddlmZmZddlmZddlmZGddeZ Gdd Z y) N) BaseTrack TrackState)matchingKalmanFilterXYAHceZdZdZeZdZdZedZ ee jddfdZ dZ dd Zd Zd Zed Zed ZedZedZedZdZy)STracka Single object tracking representation that uses Kalman filtering for state estimation. This class is responsible for storing all the information regarding individual tracklets and performs state updates and predictions based on Kalman filter. Attributes: shared_kalman (KalmanFilterXYAH): Shared Kalman filter that is used across all STrack instances for prediction. _tlwh (np.ndarray): Private attribute to store top-left corner coordinates and width and height of bounding box. kalman_filter (KalmanFilterXYAH): Instance of Kalman filter used for this particular object track. mean (np.ndarray): Mean state estimate vector. covariance (np.ndarray): Covariance of state estimate. is_activated (bool): Boolean flag indicating if the track has been activated. score (float): Confidence score of the track. tracklet_len (int): Length of the tracklet. cls (any): Class label for the object. idx (int): Index or identifier for the object. frame_id (int): Current frame ID. start_frame (int): Frame where the object was first detected. Methods: predict(): Predict the next state of the object using Kalman filter. multi_predict(stracks): Predict the next states for multiple tracks. multi_gmc(stracks, H): Update multiple track states using a homography matrix. activate(kalman_filter, frame_id): Activate a new tracklet. re_activate(new_track, frame_id, new_id): Reactivate a previously lost tracklet. update(new_track, frame_id): Update the state of a matched track. convert_coords(tlwh): Convert bounding box to x-y-angle-height format. tlwh_to_xyah(tlwh): Convert tlwh bounding box to xyah format. tlbr_to_tlwh(tlbr): Convert tlbr bounding box to tlwh format. tlwh_to_tlbr(tlwh): Convert tlwh bounding box to tlbr format. ctj|j|ddtj|_d|_d\|_|_d|_||_ d|_ ||_ |d|_ y)zInitialize new STrack instance.Ndtype)NNFr) npasarray tlbr_to_tlwhfloat32_tlwh kalman_filtermean covariance is_activatedscore tracklet_lenclsidx)selftlwhrrs lC:\Users\daisl\Desktop\realtime-object-detection\venv\Lib\site-packages\ultralytics/trackers/byte_tracker.py__init__zSTrack.__init__.shZZ 1 1$s) Ec*||_|j|_|jj|j |j \|_|_d|_tj|_ |dk(rd|_ ||_ ||_y)zStart a new tracklet.rrTN)rnext_idtrack_idinitiateconvert_coordsrrrrrr%r$rframe_id start_frame)rrrGs ractivatezSTrack.activatecs{*  %)%7%7%@%@ATATUYU_U_A`%a" 4?'' q= $D   #r c|jj|j|j|j |j \|_|_d|_tj|_ d|_ ||_ |r|j|_ |j|_|j|_|j |_y)z9Reactivates a previously lost track with a new detection.rTN)rupdaterrrFrrrr%r$rrGrCrDrrr)r new_trackrGnew_ids r re_activatezSTrack.re_activateps%)%7%7%>%>tyy$//?C?R?RS\SaSa?b&d" 4?''     LLNDM__ ====r c||_|xjdz c_|j}|jj |j |j |j|\|_|_tj|_ d|_ |j|_ |j|_ |j|_y)z Update the state of a matched track. Args: new_track (STrack): The new track containing updated information. frame_id (int): The ID of the current frame. rTN)rGrrrrKrrrFrr%r$rrrr)rrLrGnew_tlwhs rrKz STrack.update~s!  Q>>%)%7%7%>%>tyy$//?C?R?RS[?\&^" 4?''  __ ====r c$|j|S)zYConvert a bounding box's top-left-width-height format to its x-y-angle-height equivalent.) tlwh_to_xyah)rrs rrFzSTrack.convert_coordss  &&r c|j|jjS|jddj}|dxx|dzcc<|ddxxx|dddz zccc|S)zTGet current position in bounding box format (top left x, top left y, width, height).Nr7r4r5)rrr#rrets rrz STrack.tlwhsh 99 ::??$ $iim  " A#a& BQ3qr7Q; r c\|jj}|ddxxx|ddz ccc|S)z\Convert bounding box to format (min x, min y, max x, max y), i.e., (top left, bottom right).r4N)rr#rTs rtlbrz STrack.tlbrs/iinn AB3r7 r ctj|j}|ddxxx|dddz z ccc|dxx|dzcc<|S)zConvert bounding box to format (center x, center y, aspect ratio, height), where the aspect ratio is width / height. Nr4r5rrr#rrUs rrRzSTrack.tlwh_to_xyahsL jj##% BQ3qr7Q; A#a& r cntj|j}|ddxxx|ddzccc|S)zFConverts top-left bottom-right format to top-left width height format.r4NrY)rWrUs rrzSTrack.tlbr_to_tlwh6jj##% AB3r7 r cntj|j}|ddxxx|ddz ccc|S)z1Converts tlwh bounding box format to tlbr format.r4NrYrZs r tlwh_to_tlbrzSTrack.tlwh_to_tlbrr\r cVd|jd|jd|jdS)z`Return a string representation of the BYTETracker object with start and end frames and track ID.OT_z_(-))rDrH end_framers r__repr__zSTrack.__repr__s-T]]O2d&6&6%7q8HJJr N)F)__name__ __module__ __qualname____doc__rr+rr& staticmethodr,rr9rArIrNrKrFpropertyrrWrRrr^rer rr r sB%&M ] ( (#RVVAq\,,$ $ !!*'   Kr r cpeZdZdZd dZddZdZddZdZdZ d Z e d Z e d Z e d Zy) BYTETrackeraR BYTETracker: A tracking algorithm built on top of YOLOv8 for object detection and tracking. The class is responsible for initializing, updating, and managing the tracks for detected objects in a video sequence. It maintains the state of tracked, lost, and removed tracks over frames, utilizes Kalman filtering for predicting the new object locations, and performs data association. Attributes: tracked_stracks (list[STrack]): List of successfully activated tracks. lost_stracks (list[STrack]): List of lost tracks. removed_stracks (list[STrack]): List of removed tracks. frame_id (int): The current frame ID. args (namespace): Command-line arguments. max_time_lost (int): The maximum frames for a track to be considered as 'lost'. kalman_filter (object): Kalman Filter object. Methods: update(results, img=None): Updates object tracker with new detections. get_kalmanfilter(): Returns a Kalman filter object for tracking bounding boxes. init_track(dets, scores, cls, img=None): Initialize object tracking with detections. get_dists(tracks, detections): Calculates the distance between tracks and detections. multi_predict(tracks): Predicts the location of tracks. reset_id(): Resets the ID counter of STrack. joint_stracks(tlista, tlistb): Combines two lists of stracks. sub_stracks(tlista, tlistb): Filters out the stracks present in the second list from the first list. remove_duplicate_stracks(stracksa, stracksb): Removes duplicate stracks based on IOU. cg|_g|_g|_d|_||_t |dz |j z|_|j|_ |jy)zPInitialize a YOLOv8 object to track objects with given arguments and frame rate.rg>@N) tracked_stracks lost_stracksremoved_stracksrGargsint track_buffer max_time_lostget_kalmanfilterrreset_id)rrs frame_rates rrzBYTETracker.__init__s_!!   d!2T5F5F!FG!224 r Nc |xjdz c_g}g}g}g}|j}|j}tj|tj t |jddgd}|j} ||jjkD} ||jjkD} ||jjk} tj| | } || }|| }|| }|| }| | }| | }|j||||}g}g}|jD]1}|js|j!|!|j!|3|j#||j$}|j'|t)|drJ|H|j*j-||}t.j1||t.j1|||j3||}t5j6||jj8\}}}|D]\}}||}||} |j:t<j>k(r.|jA| |j|j!|[|jC| |jd|j!||j||||}!|D"cgc](}"||"j:t<j>k(s$||"*}#}"t5jD|#|!}t5j6|d \}}}$|D]\}}|#|}|!|} |j:t<j>k(r.|jA| |j|j!|[|jC| |jd|j!||D]F}%|#|%}|j:t<jFk7s&|jI|j!|H|D"cgc]}"||" }}"|j3||}t5j6|d \}}&}|D];\}}||jA|||j|j!||=|&D](}%||%}|jK|j!|*|D]b}'||'}|jL|jjNkr,|jQ|jR|j|j!|d|j$D]J}|j|jTz |jVkDs*|jK|j!|L|jD(cgc]"}(|(j:t<j>k(s!|($c}(|_|j#|j||_|j#|j||_|jY|j$|j|_|j$j[||jY|j$|j\|_|j_|j|j$\|_|_|j\j[|t |j\d kDr|j\d d|_.tj`|jD)cgc]X})|)js|)jbje|)jf|)jL|)j|)jhgzZc})tjj Scc}"wcc}"wcc}(wcc})w)zUUpdates object tracker with new detections and returns tracked object bounding boxes.rr )axisgmcN)threshF)rMg?gffffff?iir )6rGconfxyxyr concatenatearanger)reshaperrstrack_high_threshtrack_low_thresh logical_and init_trackrprappend joint_stracksrqr,hasattrr|applyr rA get_distsrlinear_assignment match_threshr$rr%rKrN iou_distanceLost mark_lost mark_removedrnew_track_threshrIrrcrv sub_stracksextendrrremove_duplicate_stracksrrWtolistrDrr)*rresultsimgactivated_stracksrefind_stracksrqrrscoresbboxesr remain_indsinds_low inds_high inds_second dets_seconddets scores_keep scores_secondcls_keep cls_second detections unconfirmedrptrack strack_poolwarpdistsmatchesu_track u_detectionitrackedidetdetdetections_secondr2r_tracked_stracksu_detection_secondit u_unconfirmedinewr@xs* rrKzBYTETracker.updates   3v;)?)G)GA)N OVXYkktyy::: DII666TYY888 nnXy9 [) k"[) {+ {#% __T;#F  ))E%%""5)&&u- * (($:K:KL  ;' 4 CO88>>#t,D   [$ /   [$ /{J7(0(B(B5QUQZQZQgQg(h%+%NHd)ET"C{{j000 S$--0!((/!!#t}}U!C%%e,&!OOK TWX5<kW A@T@TXbXjXj@j[^Wk%%&79JK/7/I/I%X[/\,,%NHd%h/E#D)C{{j000 S$--0!((/!!#t}}U!C%%e,&B%b)E{{joo-!##E*  .99[jm[ 9{J7.6.H.HWZ.[+ %NHd  ! ( (D)94== I  $ $[%: ;& BOE     " "5 )  Dt$E{{TYY777 NN4--t}} =  $ $U +  &&E}}u.1C1CC""$&&u-' ,0+?+?a+?a177jN`N`C`+?a#11$2F2FHYZ#11$2F2FW ,,T->->@T@TU   . ,,T->->@T@TU262O2OPTPdPdfjfwfw2x/d/ ##O4 t## $t +#'#7#7#>D zzLPL`L` sL`qdedrdrQVV]]_ AGGQUUAEEB BL` s** kl(:. b ts+%](.]( ]-;"]2]24]7A ]7ctS)z;Returns a Kalman filter object for tracking bounding boxes.rrds rrwzBYTETracker.get_kalmanfilteras !!r c t|r.t|||Dcgc]\}}}t|||c}}}SgScc}}}w)zMInitialize object tracking with detections and scores using STrack algorithm.)r)r-r )rrrrrrscs rrzBYTETracker.init_trackesBRUVZR[s47MN7M|atQ"7MNcaccNs<c^tj||}tj||}|S)zQCalculates the distance between tracks and detections using IOU and fuses scores.)rr fuse_score)rtracksrrs rrzBYTETracker.get_distsis-%%fj9##E:6 r c.tj|y)z6Returns the predicted tracks using the YOLOv8 network.N)r r,)rrs rr,zBYTETracker.multi_predictqsV$r c,tjy)z Resets the ID counter of STrack.N)r rxrds rrxzBYTETracker.reset_idus r ci}g}|D]"}d||j<|j|$|D]7}|j}|j|dr"d||<|j|9|S)z/Combine two lists of stracks into a single one.rr)rDrget)tlistatlistbexistsresr@tids rrzBYTETracker.joint_stracksysoA!"F1::  JJqMA**C::c1%s  1   r c|Dchc]}|j}}|Dcgc]}|j|vs|c}Scc}wcc}w)a$DEPRECATED CODE in https://github.com/ultralytics/ultralytics/pull/1890/ stracks = {t.track_id: t for t in tlista} for t in tlistb: tid = t.track_id if stracks.get(tid, 0): del stracks[tid] return list(stracks.values()) )rD)rrr@ track_ids_bs rrzBYTETracker.sub_strackssD,226aqzz6 2!C6aQZZ{%B6CC3Cs 9>>ctj||}tj|dk}gg}}t |D]k\}}||j ||j z }||j ||j z } || kDr|j|[|j|mt|D cgc] \} } | |vs | } } } t|D cgc] \} } | |vs | } } } | | fScc} } wcc} } w)z7Remove duplicate stracks with non-maximum IOU distance.g333333?) rrrwherer-rGrHrr*)stracksastracksbpdistpairsdupadupbpqtimeptimeqr2r@resaresbs rrz$BYTETracker.remove_duplicate_strackss%%h9&dKDAqQK((8A;+B+BBEQK((8A;+B+BBEu} A A  (1C1daQd]1C'1C1daQd]1CTzDCs7 C1C1 C7&C7))N)rfrgrhrirrKrwrrr,rxrjrrrrlr rrnrnsk8 qf"d%   D Dr rn) numpyr basetrackrrutilsrutils.kalman_filterrr rnrlr rrs0,1xKYxKv__r