Slicing Aided Hyper Inference and Fine-tuning for Small Object Detection
About
Detection of small objects and objects far away in the scene is a major challenge in surveillance applications. Such objects are represented by small number of pixels in the image and lack sufficient details, making them difficult to detect using conventional detectors. In this work, an open-source framework called Slicing Aided Hyper Inference (SAHI) is proposed that provides a generic slicing aided inference and fine-tuning pipeline for small object detection. The proposed technique is generic in the sense that it can be applied on top of any available object detector without any fine-tuning. Experimental evaluations, using object detection baselines on the Visdrone and xView aerial object detection datasets show that the proposed inference method can increase object detection AP by 6.8%, 5.1% and 5.3% for FCOS, VFNet and TOOD detectors, respectively. Moreover, the detection accuracy can be further increased with a slicing aided fine-tuning, resulting in a cumulative increase of 12.7%, 13.4% and 14.5% AP in the same order. Proposed technique has been integrated with Detectron2, MMDetection and YOLOv5 models and it is publicly available at https://github.com/obss/sahi.git .
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Object Detection | VisDrone (val) | AP5043.5 | 155 | |
| Object Detection | UAVNet-MS (test) | AP5015.7 | 23 | |
| Object Counting | LVIS-Count 100-image zero-shot (test) | MAE46.57 | 13 | |
| Object Detection | DOTA v1.5 | mAP5051.8 | 12 | |
| Object Detection | VisDrone | mAP5038.8 | 12 | |
| Object Detection | VOC 2012 (val) | mAP5081.8 | 8 | |
| Object Detection | PCB-Defect D1 | mAP@5072 | 5 | |
| Object Detection | HRIPCB D2 | mAP@5093.5 | 5 |