Papers › ARTrackV2: Prompting Autoregressive Tracker Where to Look and How to Describe

ARTrackV2: Prompting Autoregressive Tracker Where to Look and How to Describe

28 Dec 2023CVPR 2024 1arXiv:2312.17133archive 2025-07-28

Yifan Bai, Zeyang Zhao, Yihong Gong, Xing Wei

We present ARTrackV2, which integrates two pivotal aspects of tracking: determining where to look (localization) and how to describe (appearance analysis) the target object across video frames. Building on the foundation of its predecessor, ARTrackV2 extends the concept by introducing a unified generative framework to "read out" object's trajectory and "retell" its appearance in an autoregressive manner. This approach fosters a time-continuous methodology that models the joint evolution of motion and visual features, guided by previous estimates. Furthermore, ARTrackV2 stands out for its efficiency and simplicity, obviating the less efficient intra-frame autoregression and hand-tuned parameters for appearance updates. Despite its simplicity, ARTrackV2 achieves state-of-the-art performance on prevailing benchmark datasets while demonstrating remarkable efficiency improvement. In particular, ARTrackV2 achieves AO score of 79.5\% on GOT-10k, and AUC of 86.1\% on TrackingNet while being 3.6 × faster than ARTrack. The code will be released.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

miv-xjtu/artrack officialpytorchApache-2.0 report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

ObjectObject TrackingTemplate MatchingVideo Object TrackingVisual Object Tracking

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Visual Object Tracking GOT-10k ARTrackV2-L Average Overlap 79.5 #7 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k ARTrackV2-L Success Rate 0.5 87.8 #7 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k ARTrackV2-L Success Rate 0.75 79.6 #7 of 42 Archive leaderboard report
Visual Object Tracking LaSOT ARTrackV2-L AUC 73.6 #12 of 46 Archive leaderboard report
Visual Object Tracking LaSOT ARTrackV2-L Normalized Precision 82.8 #12 of 46 Archive leaderboard report
Visual Object Tracking LaSOT ARTrackV2-L Precision 81.1 #12 of 46 Archive leaderboard report
Visual Object Tracking LaSOT-ext ARTrackV2-L AUC 53.4 #10 of 18 Archive leaderboard report
Visual Object Tracking LaSOT-ext ARTrackV2-L Normalized Precision 63.7 #10 of 18 Archive leaderboard report
Visual Object Tracking LaSOT-ext ARTrackV2-L Precision 60.2 #10 of 18 Archive leaderboard report
Visual Object Tracking NeedForSpeed ARTrackV2-L AUC 0.684 #2 of 10 Archive leaderboard report
Visual Object Tracking TNL2K ARTrackV2-L AUC 61.6 #9 of 16 Archive leaderboard report
Visual Object Tracking TrackingNet ARTrackV2-L Accuracy 86.1 #5 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet ARTrackV2-L Normalized Precision 90.4 #5 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet ARTrackV2-L Precision 86.2 #5 of 40 Archive leaderboard report
Visual Object Tracking UAV123 ARTrackV2-L AUC 0.717 #4 of 16 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AO

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections