FSD: Fully-Specialized Detector via Neural Architecture Search
This work addresses the need for task-specific object detectors in domains like medical imaging, offering an automated solution to reduce manual tuning, though it is incremental as it builds on existing neural architecture search techniques.
The paper tackles the problem of designing efficient and effective object detectors for specialized domains, such as medical lesion detection, by proposing a fully-automatic pipeline using neural architecture search. It achieves a 3.1 mAP gain with 40% fewer parameters on binary lesion detection and a 10% mAP improvement on multi-type lesion detection compared to existing methods.
Most generic object detectors are mainly built for standard object detection tasks such as COCO and PASCAL VOC. They might not work well and/or efficiently on tasks of other domains consisting of images that are visually different from standard datasets. To this end, many advances have been focused on adapting a general-purposed object detector with limited domain-specific designs. However, designing a successful task-specific detector requires extraneous manual experiments and parameter tuning through trial and error. In this paper, we first propose and examine a fully-automatic pipeline to design a fully-specialized detector (FSD) which mainly incorporates a neural-architectural-searched model by exploring ideal network structures over the backbone and task-specific head. On the DeepLesion dataset, extensive results show that FSD can achieve 3.1 mAP gain while using approximately 40% fewer parameters on binary lesion detection task and improved the mAP by around 10% on multi-type lesion detection task via our region-aware graph modeling compared with existing general-purposed medical lesion detection networks.