MDSSD: Multi-scale Deconvolutional Single Shot Detector for Small Objects
This work addresses the challenge of detecting small objects in computer vision, which is critical for applications like surveillance and autonomous driving, and represents an incremental improvement over existing multi-scale detectors.
The paper tackles the problem of small object detection by designing MDSSD, a multi-scale deconvolutional single shot detector that upsamples high-level feature maps and fuses them with low-level features to enhance semantic information, achieving 77.6% mAP on TT100K and 2-5 point improvements on small object categories in PASCAL VOC2007 and MS COCO.
For most of the object detectors based on multi-scale feature maps, the shallow layers are rich in fine spatial information and thus mainly responsible for small object detection. The performance of small object detection, however, is still less than satisfactory because of the deficiency of semantic information on shallow feature maps. In this paper, we design a Multi-scale Deconvolutional Single Shot Detector (MDSSD), especially for small object detection. In MDSSD, multiple high-level feature maps at different scales are upsampled simultaneously to increase the spatial resolution. Afterwards, we implement the skip connections with low-level feature maps via Fusion Block. The fusion feature maps, named Fusion Module, are of strong feature representational power of small instances. It is noteworthy that these high-level feature maps utilized in Fusion Block preserve both strong semantic information and some fine details of small instances, rather than the top-most layer where the representation of fine details for small objects are potentially wiped out. The proposed framework achieves 77.6% mAP for small object detection on the challenging dataset TT100K with 512 x 512 input, outperforming other detectors with a large margin. Moreover, it can also achieve state-of-the-art results for general object detection on PASCAL VOC2007 test and MS COCO test-dev2015, especially achieving 2 to 5 points improvement on small object categories.