paper-with-me

Multi-Object Tracking 벤치마크

Multi-Object Tracking on DanceTrack

80개 결과 · ⬇ CSV · JSON

HOTA

39.7 48.75 57.8 66.85 75.9 2020-04 2026-11 CenterTrack — 41.8 (2020-04-02) CenterTrack — 41.8 (2020-04-02) FairMOT — 39.7 (2020-04-04) FairMOT — 39.7 (2020-04-04) QDTrack — 45.7 (2020-06-11) QDTrack — 45.7 (2020-06-11) TransTrack — 45.7 (2020-12-31) TransTrack — 45.7 (2020-12-31) TraDes — 43.3 (2021-03-16) TraDes — 43.3 (2021-03-16) MOTR — 54.2 (2021-05-07) MOTR — 54.2 (2021-05-07) ByteTrack — 47.1 (2021-10-13) ByteTrack — 47.1 (2021-10-13) OC-SORT — 55.1 (2022-03-27) OC-SORT — 55.1 (2022-03-27) FCG — 48.7 (2022-10-07) FCG — 48.7 (2022-10-07) MOTRv2 — 73.4 (2022-11-17) MOTRv2 — 73.4 (2022-11-17) C-BIoU — 60.6 (2022-11-24) C-BIoU — 60.6 (2022-11-24) MT_IOT — 66.66 (2022-12-07) MT_IOT — 66.66 (2022-12-07) Deep OC-SORT — 61.3 (2023-02-23) Deep OC-SORT — 61.3 (2023-02-23) CO-MOT — 69.4 (2023-05-22) CO-MOT — 69.4 (2023-05-22) MotionTrack — 58.2 (2023-06-05) MotionTrack — 58.2 (2023-06-05) SparseTrack — 55.7 (2023-06-08) SparseTrack — 55.7 (2023-06-08) MeMOTR — 68.5 (2023-07-28) MeMOTR (Deformable DETR) — 63.4 (2023-07-28) MeMOTR — 68.5 (2023-07-28) MeMOTR (Deformable DETR) — 63.4 (2023-07-28) Hybrid-SORT-ReID — 65.7 (2023-08-01) Hybrid-SORT — 62.2 (2023-08-01) Hybrid-SORT-ReID — 65.7 (2023-08-01) Hybrid-SORT — 62.2 (2023-08-01) UCMCTrack — 63.6 (2023-12-14) UCMCTrack — 63.6 (2023-12-14) MoveSORT — 56.1 (2024-02-15) MoveSORT — 56.1 (2024-02-15) C-TWiX — 62.1 (2024-03-12) C-TWiX — 62.1 (2024-03-12) MambaMOT — 56.1 (2024-03-16) MambaMOT — 56.1 (2024-03-16) MOTIP (Deformable DETR, with DanceTrack val and CrowdHuman) — 73.7 (2024-03-25) MOTIP (Deformable DETR, with CrowdHuman) — 71.4 (2024-03-25) MOTIP (DAB-Deformable DETR) — 70.0 (2024-03-25) MOTIP (Deformable DETR) — 67.5 (2024-03-25) MOTIP (Deformable DETR, with DanceTrack val and CrowdHuman) — 73.7 (2024-03-25) MOTIP (Deformable DETR, with CrowdHuman) — 71.4 (2024-03-25) MOTIP (DAB-Deformable DETR) — 70.0 (2024-03-25) MOTIP (Deformable DETR) — 67.5 (2024-03-25) ETTrack — 56.4 (2024-05-24) ETTrack — 56.4 (2024-05-24) DeepMoveSORT — 63.0 (2024-06-30) DeepMoveSORT — 63.0 (2024-06-30) IMM-JHSE — 66.24 (2024-09-04) IMM-JHSE — 66.24 (2024-09-04) Fast-StrongSORT — 55.9 (2024-09-10) Fast-StrongSORT — 55.9 (2024-09-10) AED — 66.6 (2024-09-14) AED — 66.6 (2024-09-14) CMTrack — 61.8 (2024-09-27) CMTrack — 61.8 (2024-09-27) SAM2MOT — 75.9 (2025-04-06) SAM2MOT — 75.9 (2025-04-06) CAMELTrack (fully online) — 69.3 (2025-05-02) CAMELTrack (fully online) — 69.3 (2025-05-02) TrackTrack — 66.5 (2025-06-15) TrackTrack — 66.5 (2025-06-15) MeMoSORT — 67.9 (2025-08-13) Motion-Aware — 72.2 (2025-09-26) SelfMOTR — 69.2 (2025-11-25) Occlusion-Aware — 63.1 (2026-03-06) SpikingMOT — 56.5 (2026-07-22) CenterTrack — 41.8 (2020-04-02) QDTrack — 45.7 (2020-06-11) MOTR — 54.2 (2021-05-07) OC-SORT — 55.1 (2022-03-27) MOTRv2 — 73.4 (2022-11-17) MOTIP (Deformable DETR, with DanceTrack val and CrowdHuman) — 73.7 (2024-03-25) SAM2MOT — 75.9 (2025-04-06)
RankModel HOTAMOTAIDF1AssADetA Extra Training Data PaperCodeYear
1 SAM2MOT 75.988.984.472.379.9 SAM2MOT: A Novel Paradigm of Multi-Object Tracking by Segmentation TripleJoy/SAM2MOT 2025
2 MOTIP (Deformable DETR, with DanceTrack val and CrowdHuman) 73.792.778.465.982.6 ✓ Multiple Object Tracking as ID Prediction MCG-NJU/MOTIP 2024
3 MOTRv2 73.492.176.064.483.7 ✓ MOTRv2: Bootstrapping End-to-End Multi-Object Tracking by Pretrained Object Detectors megvii-research/MOTRv2 · DanceTrack/DanceTrack · bingfengyan/visam · +1 2022
4 Motion-Aware 자동 추출 72.2–––– Motion-Aware Transformer for Multi-Object Tracking 2025
5 MOTIP (Deformable DETR, with CrowdHuman) 71.491.676.362.881.3 ✓ Multiple Object Tracking as ID Prediction MCG-NJU/MOTIP 2024
6 MOTIP (DAB-Deformable DETR) 70.091.075.160.880.8 Multiple Object Tracking as ID Prediction MCG-NJU/MOTIP 2024
7 CO-MOT 69.491.271.958.982.1 ✓ Bridging the Gap Between End-to-end and Non-End-to-end Multi-Object Tracking bingfengyan/visam · BingfengYan/CO-MOT 2023
8 CAMELTrack (fully online) 69.3–––– CAMELTrack: Context-Aware Multi-cue ExpLoitation for Online Multi-Object Tracking TrackingLaboratory/CAMELTrack 2025
9 SelfMOTR 자동 추출 69.2–––– SelfMOTR: Revisiting MOTR with Self-Generating Detection Priors 2025
10 MeMOTR 68.589.971.258.480.5 MeMOTR: Long-Term Memory-Augmented Transformer for Multi-Object Tracking mcg-nju/memotr 2023
11 MeMoSORT 자동 추출 67.9–––– MeMoSORT: Memory-Assisted Filtering and Motion-Adaptive Association Metric for Multi-Person Tracking 2025
12 MOTIP (Deformable DETR) 67.590.372.257.679.4 Multiple Object Tracking as ID Prediction MCG-NJU/MOTIP 2024
13 MT_IOT 66.6693.9770.652.9584.14 ✓ Multiple Object Tracking Challenge Technical Report for Team MT_IoT BingfengYan/DS_OCSORT 2022
14 AED 66.692.269.754.382.0 ✓ Associate Everything Detected: Facilitating Tracking-by-Detection to the Unknown balabooooo/aed 2024
15 TrackTrack 66.593.667.852.9– ✓ Focusing on Tracks for Online Multi-Object Tracking kamkyu94/TrackTrack 2025
16 IMM-JHSE 66.2489.9571.7255.41– One Homography is All You Need: IMM-based Joint Homography and Multiple Object State Estimation Paulkie99/imm-jhse 2024
17 Hybrid-SORT-ReID 65.791.867.452.682.2 Hybrid-SORT: Weak Cues Matter for Online Multi-Object Tracking mikel-brostrom/boxmot · ymzis69/HybridSORT 2023
18 UCMCTrack 63.6–65.0–– UCMCTrack: Multi-Object Tracking with Uniform Camera Motion Compensation corfyi/ucmctrack 2023
19 MeMOTR (Deformable DETR) 63.485.465.552.377.0 MeMOTR: Long-Term Memory-Augmented Transformer for Multi-Object Tracking mcg-nju/memotr 2023
20 Occlusion-Aware 자동 추출 63.1–––– Occlusion-Aware SORT: Observing Occlusion for Robust Multi-Object Tracking 2026
1–20 / 80 다음 → 페이지당 10 20 50 100