Feature Selective Anchor-Free Module for Single-Shot Object Detection

CVPR 2019  ·  Chenchen Zhu, Yihui He, Marios Savvides ·

We motivate and present feature selective anchor-free (FSAF) module, a simple and effective building block for single-shot object detectors. It can be plugged into single-shot detectors with feature pyramid structure... The FSAF module addresses two limitations brought up by the conventional anchor-based detection: 1) heuristic-guided feature selection; 2) overlap-based anchor sampling. The general concept of the FSAF module is online feature selection applied to the training of multi-level anchor-free branches. Specifically, an anchor-free branch is attached to each level of the feature pyramid, allowing box encoding and decoding in the anchor-free manner at an arbitrary level. During training, we dynamically assign each instance to the most suitable feature level. At the time of inference, the FSAF module can work jointly with anchor-based branches by outputting predictions in parallel. We instantiate this concept with simple implementations of anchor-free branches and online feature selection strategy. Experimental results on the COCO detection track show that our FSAF module performs better than anchor-based counterparts while being faster. When working jointly with anchor-based branches, the FSAF module robustly improves the baseline RetinaNet by a large margin under various settings, while introducing nearly free inference overhead. And the resulting best model can achieve a state-of-the-art 44.6% mAP, outperforming all existing single-shot detectors on COCO. read more

PDF Abstract CVPR 2019 PDF CVPR 2019 Abstract

Datasets


Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Object Detection COCO minival FSAF (ResNeXt-101, anchor-based branches) box AP 41.6 # 85
AP50 62.4 # 44
Object Detection COCO minival FSAF (ResNet-101, anchor-based branches) box AP 39.3 # 110
AP50 59.2 # 70
Object Detection COCO minival FSAF (ResNet-101) box AP 37.9 # 122
AP50 58.0 # 78
Object Detection COCO minival FSAF (ResNet-50) box AP 35.9 # 127
AP50 55.0 # 86
AP75 37.9 # 81
APS 19.8 # 69
APM 39.6 # 69
APL 48.2 # 71
Object Detection COCO test-dev FSAF (ResNet-101, single-scale) box AP 40.9 # 135
AP50 61.5 # 110
AP75 44 # 120
APS 24 # 101
APM 44.2 # 111
APL 51.3 # 123
Object Detection COCO test-dev FSAF (ResNeXt-101, multi-scale) box AP 44.6 # 95
AP50 65.2 # 66
AP75 48.6 # 73
APS 29.7 # 49
APM 47.1 # 83
APL 54.6 # 100

Methods