Papers › Unified Sequence-to-Sequence Learning for Single- and Multi-Modal Visual Object Tracking

Unified Sequence-to-Sequence Learning for Single- and Multi-Modal Visual Object Tracking

27 Apr 2023CVPR 2023 1arXiv:2304.14394archive 2025-07-28

Xin Chen, Ben Kang, Jiawen Zhu, Dong Wang, Houwen Peng, Huchuan Lu

In this paper, we introduce a new sequence-to-sequence learning framework for RGB-based and multi-modal object tracking. First, we present SeqTrack for RGB-based tracking. It casts visual tracking as a sequence generation task, forecasting object bounding boxes in an autoregressive manner. This differs from previous trackers, which depend on the design of intricate head networks, such as classification and regression heads. SeqTrack employs a basic encoder-decoder transformer architecture. The encoder utilizes a bidirectional transformer for feature extraction, while the decoder generates bounding box sequences autoregressively using a causal transformer. The loss function is a plain cross-entropy. Second, we introduce SeqTrackv2, a unified sequence-to-sequence framework for multi-modal tracking tasks. Expanding upon SeqTrack, SeqTrackv2 integrates a unified interface for auxiliary modalities and a set of task-prompt tokens to specify the task. This enables it to manage multi-modal tracking tasks using a unified model and parameter set. This sequence learning paradigm not only simplifies the tracking framework, but also showcases superior performance across 14 challenging benchmarks spanning five single- and multi-modal tracking tasks. The code and models are available at https://github.com/chenxin-dlut/SeqTrackv2.

PaperPDFConference PDFCode

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

Code

chenxin-dlut/seqtrackv2 officialmentioned in papermentioned on GitHubpytorch 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

DecoderObjectObject TrackingRgb-T TrackingVisual Object TrackingVisual Tracking

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Rgb-T Tracking LasHeR SeqTrackv2-L384 Precision 76.7 #3 of 39 Archive leaderboard report
Rgb-T Tracking LasHeR SeqTrackv2-L384 Success 61.0 #3 of 39 Archive leaderboard report
Rgb-T Tracking LasHeR SeqTrackv2-L256 Precision 74.1 #9 of 39 Archive leaderboard report
Rgb-T Tracking LasHeR SeqTrackv2-L256 Success 58.8 #9 of 39 Archive leaderboard report
Rgb-T Tracking LasHeR SeqTrackv2-B384 Precision 71.5 #18 of 39 Archive leaderboard report
Rgb-T Tracking LasHeR SeqTrackv2-B384 Success 56.2 #18 of 39 Archive leaderboard report
Rgb-T Tracking LasHeR SeqTrackv2-B256 Precision 70.4 #21 of 39 Archive leaderboard report
Rgb-T Tracking LasHeR SeqTrackv2-B256 Success 55.8 #21 of 39 Archive leaderboard report
Rgb-T Tracking RGBT234 SeqTrackv2-L256 Precision 92.3 #3 of 42 Archive leaderboard report
Rgb-T Tracking RGBT234 SeqTrackv2-L256 Success 68.5 #3 of 42 Archive leaderboard report
Rgb-T Tracking RGBT234 SeqTrackv2-L384 Precision 91.3 #5 of 42 Archive leaderboard report
Rgb-T Tracking RGBT234 SeqTrackv2-L384 Success 68.0 #5 of 42 Archive leaderboard report
Rgb-T Tracking RGBT234 SeqTrackv2-B384 Precision 90.0 #10 of 42 Archive leaderboard report
Rgb-T Tracking RGBT234 SeqTrackv2-B384 Success 66.3 #10 of 42 Archive leaderboard report
Rgb-T Tracking RGBT234 SeqTrackv2-B256 Precision 88.0 #20 of 42 Archive leaderboard report
Rgb-T Tracking RGBT234 SeqTrackv2-B256 Success 64.7 #20 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k SeqTrack-L384 Average Overlap 74.8 #22 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k SeqTrack-L384 Success Rate 0.5 81.9 #22 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k SeqTrack-L384 Success Rate 0.75 72.2 #22 of 42 Archive leaderboard report
Visual Object Tracking LaSOT SeqTrack-L384 AUC 72.5 #18 of 46 Archive leaderboard report
Visual Object Tracking LaSOT SeqTrack-L384 Normalized Precision 81.5 #18 of 46 Archive leaderboard report
Visual Object Tracking LaSOT SeqTrack-L384 Precision 79.3 #18 of 46 Archive leaderboard report
Visual Object Tracking LaSOT-ext SeqTrack-L384 AUC 50.7 #15 of 18 Archive leaderboard report
Visual Object Tracking LaSOT-ext SeqTrack-L384 Normalized Precision 61.6 #15 of 18 Archive leaderboard report
Visual Object Tracking LaSOT-ext SeqTrack-L384 Precision 57.5 #15 of 18 Archive leaderboard report
Visual Object Tracking NeedForSpeed SeqTrack-L384 AUC 0.662 #9 of 10 Archive leaderboard report
Visual Object Tracking OTB-2015 SeqTrack-L384 AUC 0.683 #12 of 18 Archive leaderboard report
Visual Object Tracking TNL2K SeqTrack-L384 AUC 57.8 #13 of 16 Archive leaderboard report
Visual Object Tracking TrackingNet SeqTrack-L384 Accuracy 85.5 #12 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet SeqTrack-L384 Normalized Precision 89.8 #12 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet SeqTrack-L384 Precision 85.8 #12 of 40 Archive leaderboard report
Visual Object Tracking UAV123 SeqTrack-L384 AUC 0.685 #12 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.

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