MAOD: An Efficient Anchor-Free Object Detector Based on MobileDet

For real-time object detectors, accuracy and efficiency are two important considerations. In this paper, we propose a lightweight anchor-free detector, MAOD, to better balance efficiency and accuracy. Our object detector contains three components: an efficient backbone network (MobileDet), a lightweight feature pyramid structure (L-FPN) and an anchor-free per-pixel prediction method. MobileDet and L-FPN provide more accurate and faster multi-scale feature extraction. Our anchor-free per-pixel prediction method achieves efficient classification and location regression tasks. On the benchmark MS-COCO dataset, MAOD achieves 46.1% AP at the speed of 68 FPS with the input size 512 × 512. When the input size is 800 × 800, MAOD achieves 47.1% AP at the speed of 43 FPS. The fast version of MAOD (320 × 320 input size) can run at 91 PS with 43.3% AP. Compared with other state-of-the-art object detectors, our detector has similar accuracy while maintaining extremely fast inference speed. MAOD achieves an optimal efficiency-accuracy tradeoff.