AI Paper Atlas

Research map / Detection and segmentation / Tracking

TAP-Vid: A Benchmark for Tracking Any Point in a Video

Carl Doersch, Ankush Gupta, Larisa Markeeva, Adria Recasens Continente, Lucas Smaira, Yusuf Aytar, Joao Carreira, Andrew Zisserman, Yi Yang

NeurIPS 2022 · Detection and segmentation › Tracking · approach: Transformer-based video object tracking · 30 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 TAP-Net instead of the strongest reported baseline, Kubric-VFS-Likepoint tracking · TAP-Vid-KineticsAJ: 40.5 -> 46.6Improved
use TAP-Net instead of PIPspoint tracking · TAP-Vid-KubricAJ: 59.1 -> 65.4Improved
use TAP-Net instead of PIPspoint tracking · TAP-Vid-DAVISAJ: 42.0 -> 38.4Hurt
use TAP-Net instead of Kubric-VFS-Likepoint tracking · TAP-Vid-RGB-StackingAJ: 57.9 -> 59.9Improved
add limited time-shifting in the first two backbone layers (ablation)TAP-Net · TAP-Vid-KineticsAJ: 46.0 -> 46.6Improved
use time-shifting only in the first two layers rather than every layer (ablation)TAP-Net · TAP-Vid-KineticsAJ: 45.4 -> 46.6Improved
use soft argmax instead of a 2-layer MLP regressor (ablation)TAP-Net · TAP-Vid-KineticsAJ: 45.8 -> 46.6Improved
sample queries at a stride of 5 frames rather than only at the first visible frame (ablation)TAP-Net · TAP-Vid-KineticsAJ: 38.5 -> 46.6Improved
sample queries at a stride of 5 frames rather than only at the first visible frame (ablation)TAP-Net · TAP-Vid-DAVISAJ: 33.0 -> 38.4Improved
sample queries at a stride of 5 frames rather than only at the first visible frame (ablation)TAP-Net · TAP-Vid-RGB-StackingAJ: 53.5 -> 59.9Improved
add optical-flow track assistance to human annotation (ablation)annotation pipeline · Kubric simulated annotationAJ: 74.7 -> 81.4Improved
fine-tune the Kubric-trained model on KineticsTAP-Net · JHMDB valPCK@0.1: 62.3 -> 63.4Improved

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