AI Paper Atlas

Research map / Detection and segmentation / Tracking

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

Artem Zholus, Carl Doersch, Yi Yang, Skanda Koppula, Viorica Patraucean, Xu Owen He, Ignacio Rocco, Mehdi S. M. Sajjadi, Sarath Chandar, Ross Goroshin

ICCV 2025 · Detection and segmentation › Tracking · approach: Transformer-based video object tracking · 31 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use classification coordinate head instead of regression head (ablation)TAPNext-S · DAVIS query-first (small-scale ablation)average Jaccard: 44.7 -> 55.0Improved
add classification loss to regression coordinate loss (ablation)TAPNext-S · DAVIS query-first (small-scale ablation)average Jaccard: 48.1 -> 55.0Improved
add regression loss to classification-only coordinate loss (ablation)TAPNext-S · DAVIS query-first (small-scale ablation)average Jaccard: 52.7 -> 55.0Improved
use 8x8 image patch instead of 16x16 (ablation)TAPNext-S · DAVIS query-first (small-scale ablation)average Jaccard: 49.7 -> 55.0Improved
add losses after each ViT block (ablation)TAPNext-S · DAVIS query-first (small-scale ablation)average Jaccard: 50.5 -> 55.0Improved
add 2x state expansion in the SSM (ablation)TAPNext-S · DAVIS query-first (small-scale ablation)average Jaccard: 53.8 -> 55.0Improved
use temporal SSM instead of temporal attention (ablation)TAPNext-S · DAVIS full-length videosaverage Jaccard: 17.3 -> 55.0Improved
use temporal SSM instead of temporal attention (ablation)TAPNext-S · DAVIS 24-frame videosaverage Jaccard: 68.4 -> 70.0Improved
track points jointly instead of one at a time with support grids (ablation)TAPNext-B · DAVIS firstaverage Jaccard: 62.2 -> 62.4No clear effect
add image reconstruction objective on output image tokens (ablation)TAPNext · TAP-Vidtracking performanceNo clear effect
add BootsTAP-style self-training on real videosTAPNext-B · DAVIS firstaverage Jaccard: 62.4 -> 65.2Improved
TAPNext online vs CoTracker3 online latency (256 queries, H100)TAPNext-B · latencylatency ms: 80 -> 5.05Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy