google-deepmind / google-deepmind/tapnet

Verify evals on Papers with Code

Open
#166 2 comments 0 reactions 0 assignees View on GitHub
Dominant language
Jupyter Notebook
Stars
2k
Forks
192
PR merge metrics
No merged PRs in 30d

Description

Hi,

Niels here from the open-source team at Hugging Face. Congratulations on your work!

I've made the [paper](https://paperswithcode.co/paper/2604.10582) and [4 paper-native evaluations](https://paperswithcode.co/paper/2604.10582#results) available on Papers with Code.

The paper is part of the [2D Point Tracking](https://paperswithcode.co/tasks/point-tracking) task page.

The TAPNext++ result currently ranks first on [PointOdyssey v1.2 Test (13 Videos, Query-First, 256px)](https://paperswithcode.co/benchmark/pointodyssey-v1-2-test-query-first-256px?task=point-tracking&eval=25546).

The TAPNext++ results currently rank third on [TAP-Vid DAVIS (QUERY-FIRST, 256×256)](https://paperswithcode.co/benchmark/tap-vid-davis-query-first-256px?task=point-tracking&eval=25543) and [TAP-Vid RGB-Stacking (Query-First, 256×256)](https://paperswithcode.co/benchmark/tap-vid-rgb-stacking-query-first-256px?task=point-tracking&eval=25544).

Would it be possible to verify these results and let me know if any score, model name, benchmark protocol, or openness metadata should be corrected? The imported rows are tied to the paper or its official release artifacts; comparison-table baselines were not added.

You can also edit the task, methods, project page, and GitHub URL directly from the paper page using your Hugging Face account.

If you'd like to showcase the results in your repository README, you can copy these live leaderboard badges (or use the “Copy PwC badge” button in the Results section):

[![Papers with Code: SOTA on PointOdyssey v1.2 Test (13 Videos, Query-First, 256px)](https://paperswithcode.co/api/v1/papers/2604.10582/leaderboard-badge.svg?eval=25546&live=1)](https://paperswithcode.co/api/v1/papers/2604.10582/leaderboard-badge-link?eval=25546)
[![Papers with Code: #3 on TAP-Vid DAVIS (QUERY-FIRST, 256×256)](https://paperswithcode.co/api/v1/papers/2604.10582/leaderboard-badge.svg?eval=25543&live=1)](https://paperswithcode.co/api/v1/papers/2604.10582/leaderboard-badge-link?eval=25543)
[![Papers with Code: #3 on TAP-Vid RGB-Stacking (Query-First, 256×256)](https://paperswithcode.co/api/v1/papers/2604.10582/leaderboard-badge.svg?eval=25544&live=1)](https://paperswithcode.co/api/v1/papers/2604.10582/leaderboard-badge-link?eval=25544)

Kind regards,

Niels

Contributor guide

Open the contributing guide

Research direction

Open the linked Papers with Code paper and the three benchmark pages, then compare the imported TAPNext++ rows with the paper or its official release artifacts. Verify the score, model name, benchmark protocol, and openness metadata, and report or correct any discrepancies through the paper page.

Written by the indexing model from the issue text.

Assessment

Tech stack
huggingface
Domain
documentation
Issue type
Documentation
Difficulty
3/5
Estimated time
1-2 days
Activity status
Active
Clarity
Mostly clear
Newbie friendliness
55/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.