Comprehensive preprocessing and post-processing tools for common Computer Vision tasks.
| Tasks | Subtasks | Defined Format |
Visualization | Format Conversion |
Metrics | Label Merging |
Active Learning |
|---|---|---|---|---|---|---|---|
| Classification | single-label single-label-bg multi-label(binary) |
single_label single_label_bg multi_label |
- | - | ALL | ALL | Entropy |
| Detection | - | coco voc yolo GENERAL |
ALL | between ANY two types |
ALL | ALL | horizontal flip |
| Segmentation | instance semantic |
coco GENERAL |
ALL | coco2general | ALL | - | instance semantic |
- "bg" means containing background classes which the index must start from 0 in this repo.
-
SDK
pip install -e . -
Debug
docker compose build docker compose up -d
- Classification
| task | data format | compute class-0 metrics | threshold optimization |
|---|---|---|---|
| multi-class classification | single label | V | |
| anomaly classification | single label background | V (fore-back) | |
| multi-label classification | multi_label | V | V (mean) |
- Object Detection
- Develop a GENERAL format to be the most convenient.
- The formats can be summarized as following:
| format | extension | files | type | box | disadvantage |
|---|---|---|---|---|---|
| coco | .json | 1 | int | (xmin, ymin, w, h) | get label of an image |
| yolo | .txt | len(imgs) | float | (cx, cy, w/2, h/2) | visualization, compute metrics, etc. |
| voc | .xml | len(imgs) | int | (xmin, ymin, xmax, ymax) | get class list |
| general | .json | 1 | int | (xmin, ymin, xmax, ymax) | NO |
- Segmentation
- Develop a GENERAL format to be the most convenient.
- prediction format
- numpy, shape=(num_classes, H, W)
- value=0~1
| Includes | Content | Advantage |
|---|---|---|
| general.json | Includes every imgs: path, contour, filled and boxes with class | Searching |
| gt_contour_*.npy | (H, W) with {0, 1, ..., num_classes} int | Plotting |
| gt_filled_*.npy | (num_classes, H, W) with 0 or 1 int values | Compute IOU for Metrics |
| *.jpg | Raw data | - |
- Feel free to ask if you have any question.
- Notice not supported
- segmentation general2coco
- segmentation label merging
- Confusion Matrix reference here