Pith. sign in

REVIEW 3 cited by

TAPNext: Tracking Any Point (TAP) as Next Token Prediction

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2504.05579 v2 pith:Y72K42YQ submitted 2025-04-08 cs.CV

TAPNext: Tracking Any Point (TAP) as Next Token Prediction

classification cs.CV
keywords tapnexttrackingmanybiasesexistingheuristicsinductivemodel
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Tracking Any Point (TAP) in a video is a challenging computer vision problem with many demonstrated applications in robotics, video editing, and 3D reconstruction. Existing methods for TAP rely heavily on complex tracking-specific inductive biases and heuristics, limiting their generality and potential for scaling. To address these challenges, we present TAPNext, a new approach that casts TAP as sequential masked token decoding. Our model is causal, tracks in a purely online fashion, and removes tracking-specific inductive biases. This enables TAPNext to run with minimal latency, and removes the temporal windowing required by many existing state of art trackers. Despite its simplicity, TAPNext achieves a new state-of-the-art tracking performance among both online and offline trackers. Finally, we present evidence that many widely used tracking heuristics emerge naturally in TAPNext through end-to-end training. The TAPNext model and code can be found at https://tap-next.github.io/.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. LiveSVG: Zero-Shot SVG Animation via Video Generation

    cs.CV 2026-05 unverdicted novelty 7.0

    LiveSVG fits original SVG paths to a generated target video via dual-level motion (group homographies plus per-path Bezier offsets) and sphere-packing recolorization to achieve zero-shot, skeleton-free SVG animation.

  2. Learning Long-term Motion Embeddings for Efficient Kinematics Generation

    cs.CV 2026-04 unverdicted novelty 6.0

    A 64x temporally compressed motion embedding learned from trackers enables efficient conditional flow-matching generation of long-term motions that outperform video models and task-specific methods.

  3. Track and Caption Any Motion: Open-Vocabulary Spatiotemporal Captioning via Trajectory-Conditioned Generation

    cs.CV 2025-12 conditional novelty 5.0

    A trajectory-conditioned retrieval system discovers multiple motion descriptions in videos without user queries and grounds them to point tracks, evaluated mainly on MeViS.