Proposal Distribution Calibration for Few-Shot Object Detection
Code of Proposal Distribution Calibration for Few-Shot Object Detection.
The code is based on mmdetection and imTED. Thanks a lot to these outstanding work. Please refer to get_started.md and MMDET_README.md to set up the environment and prepare the data.
Data Preparation
For few-shot experiments, please refer to FsDet for data preparation. Remember to modify the path of json in the configuration files, e.g. configs/imted/few_shot_pdc/imted_faster_rcnn_vit_base_2x_base_training_coco.py.
In run_train_pdc_few_shot.sh, you can get the relevant running commands about base training stage and finetune stage for coco dataset.
More detail about experiment for voc dataset and coco dataset will release soon.
Results on VOC Benchmark
| Method | Split-1 | Split-2 | Split-3 | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Shot | 1 | 2 | 3 | 5 | 10 | 1 | 2 | 3 | 5 | 10 | 1 | 2 | 3 | 5 | 10 |
| YOLO-ft | 6.6 | 10.7 | 12.5 | 24.8 | 38.6 | 12.5 | 4.2 | 11.6 | 16.1 | 33.9 | 13.0 | 15.9 | 15.0 | 32.2 | 38.4 |
| FRCN-ft | 13.8 | 19.6 | 32.8 | 41.5 | 45.6 | 7.9 | 15.3 | 26.2 | 31.6 | 39.1 | 9.8 | 11.3 | 19.1 | 35.0 | 45.1 |
| FSRW | 14.8 | 15.5 | 26.7 | 33.9 | 47.2 | 15.7 | 15.2 | 22.7 | 30.1 | 40.5 | 21.3 | 25.6 | 28.4 | 42.8 | 45.9 |
| TFA | 39.8 | 36.1 | 44.7 | 55.7 | 56.0 | 23.5 | 26.9 | 34.1 | 35.1 | 39.1 | 30.8 | 34.8 | 42.8 | 49.5 | 49.8 |
| MPSR | 41.7 | - | 51.4 | 55.2 | 61.8 | 24.4 | - | 39.2 | 39.9 | 47.8 | 35.6 | - | 42.3 | 48.0 | 49.7 |
| CME | 41.5 | 47.5 | 50.4 | 58.2 | 60.9 | 27.2 | 30.2 | 41.4 | 42.5 | 46.8 | 34.3 | 39.6 | 45.1 | 48.3 | 51.5 |
| FSCE | 44.2 | 43.8 | 51.4 | 61.9 | 63.4 | 27.3 | 29.5 | 43.5 | 44.2 | 50.2 | 37.2 | 41.9 | 47.5 | 54.6 | 58.5 |
| imTED-S(Our Implement) | 43.4 | 51.0 | 58.1 | 67.6 | 66.6 | 23.2 | 26.9 | 39.4 | 44.2 | 52.7 | 49.9 | 48.8 | 56.4 | 61.4 | 61.1 |
| imTED-S+PDC | 50.1 | 52.3 | 60.2 | 70.7 | 68.4 | 23.3 | 28.5 | 43.2 | 48.4 | 54.6 | 53.0 | 50.8 | 57.7 | 63.8 | 62.9 |
| imTED-B(Our Implement) | 56.8 | 64.8 | 69.4 | 80.1 | 76.8 | 37.4 | 38.1 | 59.1 | 57.6 | 60.9 | 60.9 | 59.3 | 70.0 | 73.9 | 75.7 |
| imTED-B+PDC | 61.8 | 69.1 | 70.2 | 78.7 | 79.6 | 42.9 | 41.2 | 60.0 | 56.3 | 65.9 | 60.3 | 63.1 | 70.6 | 73.3 | 76.7 |
Results on COCO Benchmark
| Method | mAPnovel | |
|---|---|---|
| Shot | 10 | 30 |
| FRCN-ft | 6.5 | 11.1 |
| FSRW | 5.6 | 9.1 |
| MetaDet | 7.1 | 11.3 |
| MetaR-CNN | 8.7 | 12.4 |
| TFA | 10.0 | 13.7 |
| MPSR | 9.8 | 14.1 |
| imTED-S | 15.0 | 21.0 |
| imTED-S+PDC | 15.7 | 22.1 |
| imTED-B | 22.5 | 30.2 |
| imTED-B+PDC | 23.4 | 30.8 |
If you have any questions with our code/paper, feel free to contact me through issue or e-mail.