Point Tracking
Point Tracking, often referred to as Tracking any Point (TAP) involves acquiring, focusing on, and continuously tracking specific target point/points across video frames. The system identifies the target point, maintains focus, and predicts its movement, enabling smooth tracking even if the target moves unpredictably, or through occlusions. TAP has wide applications like object tracking, surveillance, and autonomous navigation.
Papers
Showing 1–25 of 151 papers
All datasetsTAP-Vid-DAVISPointOdysseyTAP-Vid-DAVIS-FirstTAP-Vid-KineticsTAP-Vid-Kinetics-FirstTAP-Vid-RGB-StackingPerception TestTAP-Vid
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | LocoTrack-B | Average Jaccard | 69.4 | — | Unverified |
| 2 | BootsTAPIR | Average Jaccard | 66.2 | — | Unverified |
| 3 | CoTracker | Average Jaccard | 65.9 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | LocoTrack-B | Average Jaccard | 64.8 | — | Unverified |
| 2 | CoTracker | Average Jaccard | 62.2 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | BootsTAPIR | Average Jaccard | 61.4 | — | Unverified |
| 2 | LocoTrack-B | Average Jaccard | 59.1 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | LocoTrack-B | Average Jaccard | 52.3 | — | Unverified |
| 2 | CoTracker | Average Jaccard | 48.8 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | BootsTAPIR | Average Jaccard | 72.4 | — | Unverified |
| 2 | LocoTrack-B | Average Jaccard | 70.8 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Static Baseline | Average Jaccard | 0.36 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | PIPs++ | MTE | 4.6 | — | Unverified |