Semi-Supervised Object Detection
Semi-supervised object detection uses both labeled data and unlabeled data for training. It not only reduces the annotation burden for training high-performance object detectors but also further improves the object detector by using a large number of unlabeled data.
Papers
Showing 1–25 of 115 papers
All datasetsCOCO 10% labeled dataCOCO 5% labeled dataCOCO 1% labeled dataCOCO 100% labeled dataCOCO 2% labeled dataCOCO 0.5% labeled dataCOCO
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | MixPL | mAP | 44.6 | — | Unverified |
| 2 | Semi-DETR | mAP | 43.5 | — | Unverified |
| 3 | Consistent-Teacher | mAP | 40 | — | Unverified |
| 4 | ARSL | mAP | 38.5 | — | Unverified |
| 5 | Efficient Teacher | mAP | 37.9 | — | Unverified |
| 6 | Revisiting Class Imbalance | mAP | 37.4 | — | Unverified |
| 7 | Dense Teacher | mAP | 37.13 | — | Unverified |
| 8 | MixTeacher-FCOS | mAP | 36.95 | — | Unverified |
| 9 | MixTeacher-FRCNN | mAP | 36.72 | — | Unverified |
| 10 | PseCo | mAP | 36.06 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | MixPL | mAP | 40.1 | — | Unverified |
| 2 | Semi-DETR | mAP | 40.1 | — | Unverified |
| 3 | ARSL | mAP | 34.45 | — | Unverified |
| 4 | Efficient Teacher | mAP | 34.11 | — | Unverified |
| 5 | MixTeacher-FRCNN | mAP | 34.06 | — | Unverified |
| 6 | MixTeacher-FCOS | mAP | 33.42 | — | Unverified |
| 7 | Dense Teacher | mAP | 33.01 | — | Unverified |
| 8 | PseCo | mAP | 32.5 | — | Unverified |
| 9 | Revisiting Class Imbalance | mAP | 32.21 | — | Unverified |
| 10 | VC | mAP | 32.05 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | MixPL | mAP | 31.7 | — | Unverified |
| 2 | Consistent-Teacher | mAP | 25.5 | — | Unverified |
| 3 | ARSL | mAP | 25.36 | — | Unverified |
| 4 | MixTeacher-FRCNN | mAP | 25.16 | — | Unverified |
| 5 | VC | mAP | 23.86 | — | Unverified |
| 6 | MixTeacher-FCOS | mAP | 23.83 | — | Unverified |
| 7 | Efficient Teacher | mAP | 23.76 | — | Unverified |
| 8 | PseCo | mAP | 22.43 | — | Unverified |
| 9 | MUM | mAP | 21.88 | — | Unverified |
| 10 | SSOD with OCL and RUPL | mAP | 21.63 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | MixPL | mAP | 55.2 | — | Unverified |
| 2 | Semi-DETR | mAP | 50.5 | — | Unverified |
| 3 | Consistent-Teacher | mAP | 48.2 | — | Unverified |
| 4 | Dense Teacher | mAP | 46.2 | — | Unverified |
| 5 | PseCo | mAP | 46.1 | — | Unverified |
| 6 | Soft Teacher | mAP | 44.9 | — | Unverified |
| 7 | Revisiting Class Imbalance | mAP | 44 | — | Unverified |
| 8 | RPL | mAP | 43.3 | — | Unverified |
| 9 | Adaptive Class-Rebalancing | mAP | 42.79 | — | Unverified |
| 10 | MUM | mAP | 42.11 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | MixPL | mAP | 34.7 | — | Unverified |
| 2 | Consistent-Teacher | mAP | 30.7 | — | Unverified |
| 3 | MixTeacher-FRCNN | mAP | 29.11 | — | Unverified |
| 4 | ARSL | mAP | 29.08 | — | Unverified |
| 5 | Efficient Teacher | mAP | 28.7 | — | Unverified |
| 6 | MixTeacher-FCOS | mAP | 27.88 | — | Unverified |
| 7 | PseCo | mAP | 27.77 | — | Unverified |
| 8 | VC | mAP | 27.7 | — | Unverified |
| 9 | ASTOD | mAP | 24.85 | — | Unverified |
| 10 | MUM | mAP | 24.84 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Sparse Semi-DETR | 10% | 44.3 | — | Unverified |