| SENTRY: SAM2-Enhanced Neighbor-Aware and Temporally Reasoned Memory for Visual Tracking | | 61.8 | 76.6 | 73.8 | SENTRY-D4S-L | 2026-06-23 |
| SENTRY: SAM2-Enhanced Neighbor-Aware and Temporally Reasoned Memory for Visual Tracking | | 61.5 | 75.0 | 72.9 | SENTRY-SR-L | 2026-06-23 |
| SAMURAI: Adapting Segment Anything Model for Zero-Shot Visual Tracking with Motion-Aware Memory | ✓ Link | 61.0 | 73.9 | 72.2 | SAMURAI-L | 2024-11-18 |
| A Distractor-Aware Memory for Visual Object Tracking with SAM2 | ✓ Link | 60.9 | | | DAM4SAM | 2024-11-26 |
| TR-MoE: Temporal Reliability-Aware Mixture-of-Experts for Robust Tracking | ✓ Link | 58.4 | 70.8 | 68.3 | TR-MoE | 2026-09-08 |
| SENTRY: SAM2-Enhanced Neighbor-Aware and Temporally Reasoned Memory for Visual Tracking | | 57.0 | 71.7 | 67.1 | SENTRY-S2-L | 2026-06-23 |
| Tracking Meets LoRA: Faster Training, Larger Model, Stronger Performance | ✓ Link | 56.6 | 69.0 | 65.1 | LoRAT-L-378 | 2024-03-08 |
| Tracking Meets LoRA: Faster Training, Larger Model, Stronger Performance | ✓ Link | 56.5 | 69.0 | 64.9 | LoRAT-g-378 | 2024-03-08 |
| Universal Instance Perception as Object Discovery and Retrieval | ✓ Link | 56.2 | 63.8 | 63.8 | UNINEXT-H | 2023-03-12 |
| Exploring Enhanced Contextual Information for Video-Level Object Tracking | ✓ Link | 55.7 | 66.5 | 62.9 | MCITrack-L384 | 2024-12-15 |
| RTracker: Recoverable Tracking via PN Tree Structured Memory | ✓ Link | 54.9 | 65.5 | 62.7 | RTracker-L | 2024-03-28 |
| Uni-MDTrack: Learning Decoupled Memory and Dynamic States for Parameter-Efficient Visual Tracking in All Modality | | 54.8 | 65.6 | 62.1 | Uni-MDTrack-B (RGB) | 2026-03-15 |
| Exploring Enhanced Contextual Information for Video-Level Object Tracking | ✓ Link | 54.6 | 65.7 | 62.1 | MCITrack-B224 | 2024-12-15 |
| ATCTrack: Aligning Target-Context Cues with Dynamic Target States for Robust Vision-Language Tracking | ✓ Link | 54.6 | 65.7 | 62.8 | ATCTrack-B | 2025-07-26 |
| ODONet: Online Dynamic Offset Network for Visual Object Tracking | ✓ Link | 54.3 | | | ODONet-B224 | 2026-09-08 |
| ODTrack: Online Dense Temporal Token Learning for Visual Tracking | ✓ Link | 53.9 | | | ODTrack-L | 2024-01-03 |
| ARTrackV2: Prompting Autoregressive Tracker Where to Look and How to Describe | ✓ Link | 53.4 | 63.7 | 60.2 | ARTrackV2-L | 2023-12-28 |
| HIPTrack: Visual Tracking with Historical Prompts | ✓ Link | 53.0 | 64.3 | 60.6 | HIPTrack | 2023-11-03 |
| Autoregressive Visual Tracking | ✓ Link | 52.8 | 62.9 | 59.7 | ARTrack-L | 2023-01-01 |
| DropMAE: Masked Autoencoders with Spatial-Attention Dropout for Tracking Tasks | ✓ Link | 52.7 | | 60.2 | DropTrack | 2023-04-02 |
| Autoregressive Queries for Adaptive Tracking with Spatio-TemporalTransformers | | 52.7 | 64.2 | 60.8 | AQATrack-384 (extra data) | 2024-03-15 |
| ODTrack: Online Dense Temporal Token Learning for Visual Tracking | ✓ Link | 52.4 | | | ODTrack-B | 2024-01-03 |
| Unified Sequence-to-Sequence Learning for Single- and Multi-Modal Visual Object Tracking | ✓ Link | 50.7 | 61.6 | 57.5 | SeqTrack-L384 | 2023-04-27 |
| Joint Feature Learning and Relation Modeling for Tracking: A One-Stream Framework | ✓ Link | 50.6 | 61.3 | 57.6 | OSTrack | 2022-03-22 |
| Learning Target Candidate Association to Keep Track of What Not to Track | ✓ Link | 48.2 | | | KeepTrack | 2021-03-30 |
| Adaptive Capacity Autoregressive Visual Tracking | | 47.5 | | 53.3 | ARTrack-AC_para | 2026-06-03 |
| High-Performance Long-Term Tracking with Meta-Updater | ✓ Link | 41.4 | 49.9 | 47.3 | LTMU | 2020-04-01 |