{"full_name":"mikel-brostrom/yolo_tracking","owner":"mikel-brostrom","name":"yolo_tracking","shard":"1","url":"https://github.com/mikel-brostrom/yolo_tracking","description":"BoxMOT: Pluggable Python and C++ SOTA multi-object tracking modules with support for axis-aligned and oriented bounding boxes","homepage":"https://mikel-brostrom.github.io/boxmot/","topics":["strongsort","bytetrack","ocsort","deep-learning","segmentation","tensorrt","tracking-by-detection","yolo","botsort","deepocsort","multi-object-tracking","mot","mots","multi-object-tracking-segmentation","boosttrack"],"tag_pages":["deep-learning","segmentation","yolo"],"language":"Python","license":"AGPL-3.0","avatar":"https://avatars.githubusercontent.com/u/18719680?u=468559473a17b2feadaed2b800639c83c84e19b9&v=4","created_at":"2020-06-26T09:26:23Z","pushed_at":"2026-11-12T21:03:59Z","metrics":{"stars":8326,"forks":1913,"watchers":58,"issues":1280,"prs":981,"releases":150,"issues_open":0,"issues_closed":1280,"prs_open":0,"prs_closed":402,"prs_merged":579,"discussions":88},"growth":{"day":{"stars":0,"forks":0,"base_day":"2026-11-15"},"week":{"stars":3,"forks":-3,"base_day":"2026-11-04"},"month":{"stars":12,"forks":-5,"base_day":"2026-10-17"}},"history":{"dates":["2026-09-06","2026-09-07","2026-09-08","2026-09-09","2026-09-10","2026-09-11","2026-09-12","2026-09-13","2026-09-14","2026-09-15","2026-09-16","2026-09-17","2026-09-18","2026-09-19","2026-09-20","2026-09-21","2026-09-22","2026-09-23","2026-09-24","2026-09-25","2026-09-26","2026-09-27","2026-09-28","2026-09-29","2026-09-30","2026-10-01","2026-10-02","2026-10-03","2026-10-04","2026-10-05","2026-10-06","2026-10-07","2026-10-08","2026-10-09","2026-10-10","2026-10-11","2026-10-12","2026-10-13","2026-10-14","2026-10-15","2026-10-16","2026-10-17","2026-10-18","2026-10-19","2026-10-20","2026-10-21","2026-10-22","2026-10-23","2026-10-24","2026-10-25","2026-10-26","2026-10-27","2026-10-28","2026-10-29","2026-10-30","2026-10-31","2026-11-01","2026-11-02","2026-11-03","2026-11-04","2026-11-10","2026-11-11","2026-11-12","2026-11-13","2026-11-14","2026-11-15","2026-11-16"],"stars":[8287,8287,8288,8290,8293,8293,8293,8293,8292,8293,8293,8296,8298,8299,8300,8301,8302,8302,8302,8303,8303,8302,8302,8301,8305,8304,8308,8307,8307,8309,8310,8309,8309,8310,8310,8311,8311,8311,8312,8314,8313,8314,8315,8316,8317,8319,8319,8318,8319,8319,8318,8317,8316,8320,8321,8322,8321,8322,8322,8323,8324,8325,8325,8325,8326,8326,8326],"forks":[1917,1917,1917,1918,1918,1918,1918,1919,1919,1919,1919,1919,1919,1919,1918,1917,1917,1917,1917,1917,1917,1918,1918,1918,1918,1919,1919,1919,1919,1919,1919,1919,1919,1920,1918,1918,1918,1918,1918,1918,1917,1918,1918,1918,1918,1918,1917,1917,1917,1917,1917,1917,1918,1917,1917,1917,1917,1917,1916,1916,1914,1914,1913,1913,1913,1913,1913],"watchers":[58,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,59,58,58,58,58,58,58,58,58,58,58,58,null,58,58,58,58,58,58],"issues_open":[null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,0,0,0,0,0,0],"prs_open":[null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,1,0,0,0,0,0]},"boards":{},"same_owner":[],"issues":[],"discussions":[{"number":2239,"title":"How can i use deep oc sort?","comments":1,"category":"Q&A","created_at":"2026-03-02T13:54:10Z"},{"number":2089,"title":"A question regarding a livestock multi-object tracking project in South Korea.","comments":0,"category":"Q&A","created_at":"2025-08-04T06:37:21Z"},{"number":2067,"title":"ID Not Writing To Label File","comments":4,"category":"Q&A","created_at":"2025-07-10T00:05:27Z"},{"number":2035,"title":"How do i store tracking state and them resume tracking from later","comments":0,"category":"General","created_at":"2025-06-20T07:41:30Z"},{"number":2010,"title":"Strongsort not working","comments":1,"category":"Q&A","created_at":"2025-06-06T15:07:05Z"},{"number":2006,"title":"Saving Results","comments":1,"category":"General","created_at":"2025-06-04T17:04:58Z"},{"number":1833,"title":"trackeval problem","comments":0,"category":"General","created_at":"2025-03-01T04:35:01Z"},{"number":1797,"title":"Is there any example for others trackers?","comments":0,"category":"Q&A","created_at":"2025-01-22T10:54:55Z"}],"latest_release":{"name":"Release v26.0.0","tag":"v26.0.0","published_at":"2026-10-27T11:38:17Z","body":"## BoxMOT v26.0.0\r\n\r\nBoxMOT v26.0.0 adds multimodal tracking, optional EdgeTAM mask guidance, typed tracker configuration, and joint motion-model selection. It also improves evaluation reproducibility, replay performance, and tuning diagnostics.\r\n\r\nBased on committed changes since `v25.0.0`, through `9da023fd`.\r\n\r\n### New features and improvements\r\n\r\n- **EagerMOT sensor fusion:** Track objects using calibrated 2D and 3D detections, with optional camera poses. New spatial structures represent 3D boxes, camera calibration, and multimodal outputs. Includes KITTI evaluation, class-specific tuning, and 3D Kalman calibration.\r\n- **MAF-HDA mask tracking:** Adds a Python tracker combining motion with mask-based correlation-filter appearance. Supports AABB detections with instance masks without requiring ReID weights.\r\n- **EdgeTAM integration:** Generate detection masks with the standalone segmentor or enable temporal mask-guided association using `--edgetam`. Guidance supports the nine Python box trackers with AABB geometry, IoU association, and `per_class=False`.\r\n- **Typed configuration and simpler factories:** All eleven trackers expose immutable, validated configuration classes with editor completion. Detector, ReID, and tracker factories support simpler construction, while shared `KalmanConfig` and `ReIDConfig` group component settings.\r\n- **Joint motion selection:** `tune --select-kf` compares compatible Kalman filters and noise profiles using held-out predictions. Add `--select-cmc` to jointly select camera-motion compensation. Selected profiles are saved for reuse.\r\n- **More targeted tuning:** Conditional search spaces account for geometry, association metric, and enabled features. New controls expose motion, detection-confidence, and ReID–IoU fusion, with improved trial comparisons and parameter diagnostics.\r\n- **Expanded evaluation:** Adds KITTI-MOTS mask evaluation, saved 2D detection workflows, 3D tracking metrics, and optional official KITTI AP40 reports. Materialized AABB evaluation supports ordered GTA, GSI, and GBRC postprocessing while preserving raw results.\r\n- **Faster repeated workloads:** Batched Kalman operations, optimized camera-motion compensation, indexed embedding reads, reusable workers, and optional `--cache-inputs` reduce repeated processing.\r\n- **Reliability fixes:** Restores per-sequence evaluation reporting, corrects DeepOC-SORT covariance handling during camera compensation, cleans up orphaned workers, and fixes Python 3.10 typing compatibility. OpenCV is pinned to `4.11.0.86`, and its runtime version participates in perception-cache identity.\r\n\r\n### Breaking changes\r\n\r\nThis major release includes breaking changes. When upgrading from v25:\r\n\r\n- **Update direct tracker construction.** Pass algorithm settings through the matching config, such as `BotSort(config=BotSortConfig(...))`. The factory also accepts algorithm keywords: `create_tracker(\"botsort\", match_thresh=0.8)`.\r\n- **Group component settings.** Move flat Kalman options into `kalman` and `kalman.noise`; configure tracker-owned appearance through `reid=ReIDConfig(...)` or an injected appearance encoder.\r\n- **Replace `track_buffer`.** ByteTrack and BoT-SORT now use `max_lost_frames`, counted directly in tracker updates without FPS scaling.\r\n- **Review removed tracker features.** SAM2MOT has been removed; standalone SAM segmentation remains available. OccluBoost's online GTA controls are removed, and its AABB/OBB thresholds and lifetime settings are consolidated.\r\n- **Review centroid thresholds.** `centroid_max_distance` now determines geometric eligibility across matching and recovery; overlap thresholds no longer restrict centroid matching.\r\n- **Update commands and imports.** Replace `boxmot time-variant` with `boxmot materialize --time-variant`. Internal tracker and motion modules have moved; prefer public imports from `boxmot`.\r\n- **Rebuild native libraries.** ByteTrack, BoT-SORT, and OccluBoost use constructor ABI v3. Their shared update/res","is_prerelease":false},"readme":true,"readme_zh":false,"news":[],"stale":false,"snapshot_at":"2026-11-16T09:23:14Z","history_start":"2026-09-06"}